Datasets:

bhujith10

/

multi_class_classification_dataset_openai_predictions

like 0

Title: Tangle-tree duality in abstract separation systems, Abstract: We prove a general width duality theorem for combinatorial structures with well-defined notions of cohesion and separation. These might be graphs and matroids, but can be much more general or quite different. The theorem asserts a duality between the existence of high cohesiveness somewhere local and a global overall tree structure. We describe cohesive substructures in a unified way in the format of tangles: as orientations of low-order separations satisfying certain consistency axioms. These axioms can be expressed without reference to the underlying structure, such as a graph or matroid, but just in terms of the poset of the separations themselves. This makes it possible to identify tangles, and apply our tangle-tree duality theorem, in very diverse settings. Our result implies all the classical duality theorems for width parameters in graph minor theory, such as path-width, tree-width, branch-width or rank-width. It yields new, tangle-type, duality theorems for tree-width and path-width. It implies the existence of width parameters dual to cohesive substructures such as $k$-blocks, edge-tangles, or given subsets of tangles, for which no width duality theorems were previously known. Abstract separation systems can be found also in structures quite unlike graphs and matroids. For example, our theorem can be applied to image analysis by capturing the regions of an image as tangles of separations defined as natural partitions of its set of pixels. It can be applied in big data contexts by capturing clusters as tangles. It can be applied in the social sciences, e.g. by capturing as tangles the few typical mindsets of individuals found by a survey. It could also be applied in pure mathematics, e.g. to separations of compact manifolds.

Title: Essentially No Barriers in Neural Network Energy Landscape, Abstract: Training neural networks involves finding minima of a high-dimensional non-convex loss function. Knowledge of the structure of this energy landscape is sparse. Relaxing from linear interpolations, we construct continuous paths between minima of recent neural network architectures on CIFAR10 and CIFAR100. Surprisingly, the paths are essentially flat in both the training and test landscapes. This implies that neural networks have enough capacity for structural changes, or that these changes are small between minima. Also, each minimum has at least one vanishing Hessian eigenvalue in addition to those resulting from trivial invariance.

Title: Approximating the Backbone in the Weighted Maximum Satisfiability Problem, Abstract: The weighted Maximum Satisfiability problem (weighted MAX-SAT) is a NP-hard problem with numerous applications arising in artificial intelligence. As an efficient tool for heuristic design, the backbone has been applied to heuristics design for many NP-hard problems. In this paper, we investigated the computational complexity for retrieving the backbone in weighted MAX-SAT and developed a new algorithm for solving this problem. We showed that it is intractable to retrieve the full backbone under the assumption that . Moreover, it is intractable to retrieve a fixed fraction of the backbone as well. And then we presented a backbone guided local search (BGLS) with Walksat operator for weighted MAX-SAT. BGLS consists of two phases: the first phase samples the backbone information from local optima and the backbone phase conducts local search under the guideline of backbone. Extensive experimental results on the benchmark showed that BGLS outperforms the existing heuristics in both solution quality and runtime.

Title: New type integral inequalities for convex functions with applications II, Abstract: We have recently established some integral inequalities for convex functions via the Hermite-Hadamard's inequalities. In continuation here, we also establish some interesting new integral inequalities for convex functions via the Hermite--Hadamard's inequalities and Jensen's integral inequality. Useful applications involving special means are also included.

Title: Physical problem solving: Joint planning with symbolic, geometric, and dynamic constraints, Abstract: In this paper, we present a new task that investigates how people interact with and make judgments about towers of blocks. In Experiment~1, participants in the lab solved a series of problems in which they had to re-configure three blocks from an initial to a final configuration. We recorded whether they used one hand or two hands to do so. In Experiment~2, we asked participants online to judge whether they think the person in the lab used one or two hands. The results revealed a close correspondence between participants' actions in the lab, and the mental simulations of participants online. To explain participants' actions and mental simulations, we develop a model that plans over a symbolic representation of the situation, executes the plan using a geometric solver, and checks the plan's feasibility by taking into account the physical constraints of the scene. Our model explains participants' actions and judgments to a high degree of quantitative accuracy.

Title: A Stochastic Model for File Lifetime and Security in Data Center Networks, Abstract: Data center networks are an important infrastructure in various applications of modern information technologies. Note that each data center always has a finite lifetime, thus once a data center fails, then it will lose all its storage files and useful information. For this, it is necessary to replicate and copy each important file into other data centers such that this file can increase its lifetime of staying in a data center network. In this paper, we describe a large-scale data center network with a file d-threshold policy, which is to replicate each important file into at most d-1 other data centers such that this file can maintain in the data center network under a given level of data security in the long-term. To this end, we develop three relevant Markov processes to propose two effective methods for assessing the file lifetime and data security. By using the RG-factorizations, we show that the two methods are used to be able to more effectively evaluate the file lifetime of large-scale data center networks. We hope the methodology and results given in this paper are applicable in the file lifetime study of more general data center networks with replication mechanism.

Title: A Theoretical Perspective of Solving Phaseless Compressed Sensing via Its Nonconvex Relaxation, Abstract: As a natural extension of compressive sensing and the requirement of some practical problems, Phaseless Compressed Sensing (PCS) has been introduced and studied recently. Many theoretical results have been obtained for PCS with the aid of its convex relaxation. Motivated by successful applications of nonconvex relaxed methods for solving compressive sensing, in this paper, we try to investigate PCS via its nonconvex relaxation. Specifically, we relax PCS in the real context by the corresponding $\ell_p$-minimization with $p\in (0,1)$. We show that there exists a constant $p^\ast\in (0,1]$ such that for any fixed $p\in(0, p^\ast)$, every optimal solution to the $\ell_p$-minimization also solves the concerned problem; and derive an expression of such a constant $p^\ast$ by making use of the known data and the sparsity level of the concerned problem. These provide a theoretical basis for solving this class of problems via the corresponding $\ell_p$-minimization.

Title: Affine Rough Models, Abstract: The goal of this survey article is to explain and elucidate the affine structure of recent models appearing in the rough volatility literature, and show how it leads to exponential-affine transform formulas.

Title: When flux standards go wild: white dwarfs in the age of Kepler, Abstract: White dwarf stars have been used as flux standards for decades, thanks to their staid simplicity. We have empirically tested their photometric stability by analyzing the light curves of 398 high-probability candidates and spectroscopically confirmed white dwarfs observed during the original Kepler mission and later with K2 Campaigns 0-8. We find that the vast majority (>97 per cent) of non-pulsating and apparently isolated white dwarfs are stable to better than 1 per cent in the Kepler bandpass on 1-hr to 10-d timescales, confirming that these stellar remnants are useful flux standards. From the cases that do exhibit significant variability, we caution that binarity, magnetism, and pulsations are three important attributes to rule out when establishing white dwarfs as flux standards, especially those hotter than 30,000 K.

Title: Stable representations of posets, Abstract: The purpose of this paper is to study stable representations of partially ordered sets (posets) and compare it to the well known theory for quivers. In particular, we prove that every indecomposable representation of a poset of finite type is stable with respect to some weight and construct that weight explicitly in terms of the dimension vector. We show that if a poset is primitive then Coxeter transformations preserve stable representations. When the base field is the field of complex numbers we establish the connection between the polystable representations and the unitary $\chi$-representations of posets. This connection explains the similarity of the results obtained in the series of papers.

Title: A Practical Bandit Method with Advantages in Neural Network Tuning, Abstract: Stochastic bandit algorithms can be used for challenging non-convex optimization problems. Hyperparameter tuning of neural networks is particularly challenging, necessitating new approaches. To this end, we present a method that adaptively partitions the combined space of hyperparameters, context, and training resources (e.g., total number of training iterations). By adaptively partitioning the space, the algorithm is able to focus on the portions of the hyperparameter search space that are most relevant in a practical way. By including the resources in the combined space, the method tends to use fewer training resources overall. Our experiments show that this method can surpass state-of-the-art methods in tuning neural networks on benchmark datasets. In some cases, our implementations can achieve the same levels of accuracy on benchmark datasets as existing state-of-the-art approaches while saving over 50% of our computational resources (e.g. time, training iterations).

Title: Dynamic Security Analysis of Power Systems by a Sampling-Based Algorithm, Abstract: Dynamic security analysis is an important problem of power systems on ensuring safe operation and stable power supply even when certain faults occur. No matter such faults are caused by vulnerabilities of system components, physical attacks, or cyber-attacks that are more related to cyber-security, they eventually affect the physical stability of a power system. Examples of the loss of physical stability include the Northeast blackout of 2003 in North America and the 2015 system-wide blackout in Ukraine. The nonlinear hybrid nature, that is, nonlinear continuous dynamics integrated with discrete switching, and the high degree of freedom property of power system dynamics make it challenging to conduct the dynamic security analysis. In this paper, we use the hybrid automaton model to describe the dynamics of a power system and mainly deal with the index-1 differential-algebraic equation models regarding the continuous dynamics in different discrete states. The analysis problem is formulated as a reachability problem of the associated hybrid model. A sampling-based algorithm is then proposed by integrating modeling and randomized simulation of the hybrid dynamics to search for a feasible execution connecting an initial state of the post-fault system and a target set in the desired operation mode. The proposed method enables the use of existing power system simulators for the synthesis of discrete switching and control strategies through randomized simulation. The effectiveness and performance of the proposed approach are demonstrated with an application to the dynamic security analysis of the New England 39-bus benchmark power system exhibiting hybrid dynamics. In addition to evaluating the dynamic security, the proposed method searches for a feasible strategy to ensure the dynamic security of the system in face of disruptions.

Title: Macro diversity in Cellular Networks with Random Blockages, Abstract: Blocking objects (blockages) between a transmitter and receiver cause wireless communication links to transition from line-of-sight (LOS) to non-line-of-sight (NLOS) propagation, which can greatly reduce the received power, particularly at higher frequencies such as millimeter wave (mmWave). We consider a cellular network in which a mobile user attempts to connect to two or more base stations (BSs) simultaneously, to increase the probability of at least one LOS link, which is a form of macrodiversity. We develop a framework for determining the LOS probability as a function of the number of BSs, when taking into account the correlation between blockages: for example, a single blockage close to the device -- including the user's own body -- could block multiple BSs. We consider the impact of the size of blocking objects on the system reliability probability and show that macrodiversity gains are higher when the blocking objects are small. We also show that the BS density must scale as the square of the blockage density to maintain a given level of reliability.

Title: HybridNet: Classification and Reconstruction Cooperation for Semi-Supervised Learning, Abstract: In this paper, we introduce a new model for leveraging unlabeled data to improve generalization performances of image classifiers: a two-branch encoder-decoder architecture called HybridNet. The first branch receives supervision signal and is dedicated to the extraction of invariant class-related representations. The second branch is fully unsupervised and dedicated to model information discarded by the first branch to reconstruct input data. To further support the expected behavior of our model, we propose an original training objective. It favors stability in the discriminative branch and complementarity between the learned representations in the two branches. HybridNet is able to outperform state-of-the-art results on CIFAR-10, SVHN and STL-10 in various semi-supervised settings. In addition, visualizations and ablation studies validate our contributions and the behavior of the model on both CIFAR-10 and STL-10 datasets.

Title: Multipath Error Correction in Radio Interferometric Positioning Systems, Abstract: The radio interferometric positioning system (RIPS) is an accurate node localization method featuring a novel phase-based ranging process. Multipath is the limiting error source for RIPS in ground-deployed scenarios or indoor applications. There are four distinct channels involved in the ranging process for RIPS. Multipath reflections affect both the phase and amplitude of the ranging signal for each channel. By exploiting untapped amplitude information, we put forward a scheme to estimate each channel's multipath profile, which is then subsequently used to correct corresponding errors in phase measurements. Simulations show that such a scheme is very effective in reducing multipath phase errors, which are essentially brought down to the level of receiver noise under moderate multipath conditions. It is further demonstrated that ranging errors in RIPS are also greatly reduced via the proposed scheme.

Title: Network of vertically c-oriented prism shaped InN nanowalls grown on c-GaN/sapphire template by chemical vapor deposition technique, Abstract: Networks of vertically c-oriented prism shaped InN nanowalls, are grown on c-GaN/sapphire templates using a CVD technique, where pure indium and ammonia are used as metal and nitrogen precursors. A systematic study of the growth, structural and electronic properties of these samples shows a preferential growth of the islands along [11-20] and [0001] directions leading to the formation of such a network structure, where the vertically [0001] oriented tapered walls are laterally align along one of the three [11-20] directions. Inclined facets of these walls are identified as r-planes [(1-102)-planes] of wurtzite InN. Onset of absorption for these samples is observed to be higher than the band gap of InN suggesting a high background carrier concentration in this material. Study of the valence band edge through XPS indicates the formation of positive depletion regions below the r-plane side facets of the walls. This is in contrast with the observation for c-plane InN epilayers, where electron accumulation is often reported below the top surface.

Title: Convolved subsampling estimation with applications to block bootstrap, Abstract: The block bootstrap approximates sampling distributions from dependent data by resampling data blocks. A fundamental problem is establishing its consistency for the distribution of a sample mean, as a prototypical statistic. We use a structural relationship with subsampling to characterize the bootstrap in a new and general manner. While subsampling and block bootstrap differ, the block bootstrap distribution of a sample mean equals that of a $k$-fold self-convolution of a subsampling distribution. Motivated by this, we provide simple necessary and sufficient conditions for a convolved subsampling estimator to produce a normal limit that matches the target of bootstrap estimation. These conditions may be linked to consistency properties of an original subsampling distribution, which are often obtainable under minimal assumptions. Through several examples, the results are shown to validate the block bootstrap for means under significantly weakened assumptions in many existing (and some new) dependence settings, which also addresses a standing conjecture of Politis, Romano and Wolf(1999). Beyond sample means, the convolved subsampling estimator may not match the block bootstrap, but instead provides a hybrid-resampling estimator of interest in its own right. For general statistics with normal limits, results also establish the consistency of convolved subsampling under minimal dependence conditions, including non-stationarity.

Title: Computing Simple Multiple Zeros of Polynomial Systems, Abstract: Given a polynomial system f associated with a simple multiple zero x of multiplicity {\mu}, we give a computable lower bound on the minimal distance between the simple multiple zero x and other zeros of f. If x is only given with limited accuracy, we propose a numerical criterion that f is certified to have {\mu} zeros (counting multiplicities) in a small ball around x. Furthermore, for simple double zeros and simple triple zeros whose Jacobian is of normalized form, we define modified Newton iterations and prove the quantified quadratic convergence when the starting point is close to the exact simple multiple zero. For simple multiple zeros of arbitrary multiplicity whose Jacobian matrix may not have a normalized form, we perform unitary transformations and modified Newton iterations, and prove its non-quantified quadratic convergence and its quantified convergence for simple triple zeros.

Title: Latent Geometry and Memorization in Generative Models, Abstract: It can be difficult to tell whether a trained generative model has learned to generate novel examples or has simply memorized a specific set of outputs. In published work, it is common to attempt to address this visually, for example by displaying a generated example and its nearest neighbor(s) in the training set (in, for example, the L2 metric). As any generative model induces a probability density on its output domain, we propose studying this density directly. We first study the geometry of the latent representation and generator, relate this to the output density, and then develop techniques to compute and inspect the output density. As an application, we demonstrate that "memorization" tends to a density made of delta functions concentrated on the memorized examples. We note that without first understanding the geometry, the measurement would be essentially impossible to make.

Title: Landau Collision Integral Solver with Adaptive Mesh Refinement on Emerging Architectures, Abstract: The Landau collision integral is an accurate model for the small-angle dominated Coulomb collisions in fusion plasmas. We investigate a high order accurate, fully conservative, finite element discretization of the nonlinear multi-species Landau integral with adaptive mesh refinement using the PETSc library (www.mcs.anl.gov/petsc). We develop algorithms and techniques to efficiently utilize emerging architectures with an approach that minimizes memory usage and movement and is suitable for vector processing. The Landau collision integral is vectorized with Intel AVX-512 intrinsics and the solver sustains as much as 22% of the theoretical peak flop rate of the Second Generation Intel Xeon Phi, Knights Landing, processor.

Title: Finite-time scaling at the Anderson transition for vibrations in solids, Abstract: A model in which a three-dimensional elastic medium is represented by a network of identical masses connected by springs of random strengths and allowed to vibrate only along a selected axis of the reference frame, exhibits an Anderson localization transition. To study this transition, we assume that the dynamical matrix of the network is given by a product of a sparse random matrix with real, independent, Gaussian-distributed non-zero entries and its transpose. A finite-time scaling analysis of system's response to an initial excitation allows us to estimate the critical parameters of the localization transition. The critical exponent is found to be $\nu = 1.57 \pm 0.02$ in agreement with previous studies of Anderson transition belonging to the three-dimensional orthogonal universality class.

Title: On the Status of the Measurement Problem: Recalling the Relativistic Transactional Interpretation, Abstract: In view of a resurgence of concern about the measurement problem, it is pointed out that the Relativistic Transactional Interpretation (RTI) remedies issues previously considered as drawbacks or refutations of the original TI. Specifically, once one takes into account relativistic processes that are not representable at the non-relativistic level (such as particle creation and annihilation, and virtual propagation), absorption is quantitatively defined in unambiguous physical terms. In addition, specifics of the relativistic transactional model demonstrate that the Maudlin `contingent absorber' challenge to the original TI cannot even be mounted: basic features of established relativistic field theories (in particular, the asymmetry between field sources and the bosonic fields, and the fact that slow-moving bound states, such as atoms, are not offer waves) dictate that the `slow-moving offer wave' required for the challenge scenario cannot exist. It is concluded that issues previously considered obstacles for TI are no longer legitimately viewed as such, and that reconsideration of the transactional picture is warranted in connection with solving the measurement problem.

Title: Mixed Precision Training of Convolutional Neural Networks using Integer Operations, Abstract: The state-of-the-art (SOTA) for mixed precision training is dominated by variants of low precision floating point operations, and in particular, FP16 accumulating into FP32 Micikevicius et al. (2017). On the other hand, while a lot of research has also happened in the domain of low and mixed-precision Integer training, these works either present results for non-SOTA networks (for instance only AlexNet for ImageNet-1K), or relatively small datasets (like CIFAR-10). In this work, we train state-of-the-art visual understanding neural networks on the ImageNet-1K dataset, with Integer operations on General Purpose (GP) hardware. In particular, we focus on Integer Fused-Multiply-and-Accumulate (FMA) operations which take two pairs of INT16 operands and accumulate results into an INT32 output.We propose a shared exponent representation of tensors and develop a Dynamic Fixed Point (DFP) scheme suitable for common neural network operations. The nuances of developing an efficient integer convolution kernel is examined, including methods to handle overflow of the INT32 accumulator. We implement CNN training for ResNet-50, GoogLeNet-v1, VGG-16 and AlexNet; and these networks achieve or exceed SOTA accuracy within the same number of iterations as their FP32 counterparts without any change in hyper-parameters and with a 1.8X improvement in end-to-end training throughput. To the best of our knowledge these results represent the first INT16 training results on GP hardware for ImageNet-1K dataset using SOTA CNNs and achieve highest reported accuracy using half-precision

Title: Nanostructured complex oxides as a route towards thermal behavior in artificial spin ice systems, Abstract: We have used soft x-ray photoemission electron microscopy to image the magnetization of single domain La$_{0.7}$Sr$_{0.3}$MnO$_{3}$ nano-islands arranged in geometrically frustrated configurations such as square ice and kagome ice geometries. Upon thermal randomization, ensembles of nano-islands with strong inter-island magnetic coupling relax towards low-energy configurations. Statistical analysis shows that the likelihood of ensembles falling into low-energy configurations depends strongly on the annealing temperature. Annealing to just below the Curie temperature of the ferromagnetic film (T$_{C}$ = 338 K) allows for a much greater probability of achieving low energy configurations as compared to annealing above the Curie temperature. At this thermally active temperature of 325 K, the ensemble of ferromagnetic nano-islands explore their energy landscape over time and eventually transition to lower energy states as compared to the frozen-in configurations obtained upon cooling from above the Curie temperature. Thus, this materials system allows for a facile method to systematically study thermal evolution of artificial spin ice arrays of nano-islands at temperatures modestly above room temperature.

Title: A Simple Convolutional Generative Network for Next Item Recommendation, Abstract: Convolutional Neural Networks (CNNs) have been recently introduced in the domain of session-based next item recommendation. An ordered collection of past items the user has interacted with in a session (or sequence) are embedded into a 2-dimensional latent matrix, and treated as an image. The convolution and pooling operations are then applied to the mapped item embeddings. In this paper, we first examine the typical session-based CNN recommender and show that both the generative model and network architecture are suboptimal when modeling long-range dependencies in the item sequence. To address the issues, we introduce a simple, but very effective generative model that is capable of learning high-level representation from both short- and long-range item dependencies. The network architecture of the proposed model is formed of a stack of \emph{holed} convolutional layers, which can efficiently increase the receptive fields without relying on the pooling operation. Another contribution is the effective use of residual block structure in recommender systems, which can ease the optimization for much deeper networks. The proposed generative model attains state-of-the-art accuracy with less training time in the next item recommendation task. It accordingly can be used as a powerful recommendation baseline to beat in future, especially when there are long sequences of user feedback.

Title: A Domain Specific Language for Performance Portable Molecular Dynamics Algorithms, Abstract: Developers of Molecular Dynamics (MD) codes face significant challenges when adapting existing simulation packages to new hardware. In a continuously diversifying hardware landscape it becomes increasingly difficult for scientists to be experts both in their own domain (physics/chemistry/biology) and specialists in the low level parallelisation and optimisation of their codes. To address this challenge, we describe a "Separation of Concerns" approach for the development of parallel and optimised MD codes: the science specialist writes code at a high abstraction level in a domain specific language (DSL), which is then translated into efficient computer code by a scientific programmer. In a related context, an abstraction for the solution of partial differential equations with grid based methods has recently been implemented in the (Py)OP2 library. Inspired by this approach, we develop a Python code generation system for molecular dynamics simulations on different parallel architectures, including massively parallel distributed memory systems and GPUs. We demonstrate the efficiency of the auto-generated code by studying its performance and scalability on different hardware and compare it to other state-of-the-art simulation packages. With growing data volumes the extraction of physically meaningful information from the simulation becomes increasingly challenging and requires equally efficient implementations. A particular advantage of our approach is the easy expression of such analysis algorithms. We consider two popular methods for deducing the crystalline structure of a material from the local environment of each atom, show how they can be expressed in our abstraction and implement them in the code generation framework.

Title: CMB anisotropies at all orders: the non-linear Sachs-Wolfe formula, Abstract: We obtain the non-linear generalization of the Sachs-Wolfe + integrated Sachs-Wolfe (ISW) formula describing the CMB temperature anisotropies. Our formula is valid at all orders in perturbation theory, is also valid in all gauges and includes scalar, vector and tensor modes. A direct consequence of our results is that the maps of the logarithmic temperature anisotropies are much cleaner than the usual CMB maps, because they automatically remove many secondary anisotropies. This can for instance, facilitate the search for primordial non-Gaussianity in future works. It also disentangles the non-linear ISW from other effects. Finally, we provide a method which can iteratively be used to obtain the lensing solution at the desired order.

Title: On a binary system of Prendiville: The cubic case, Abstract: We prove sharp decoupling inequalities for a class of two dimensional non-degenerate surfaces in R^5, introduced by Prendiville. As a consequence, we obtain sharp bounds on the number of integer solutions of the Diophantine systems associated with these surfaces.

Title: Causal Inference by Stochastic Complexity, Abstract: The algorithmic Markov condition states that the most likely causal direction between two random variables X and Y can be identified as that direction with the lowest Kolmogorov complexity. Due to the halting problem, however, this notion is not computable. We hence propose to do causal inference by stochastic complexity. That is, we propose to approximate Kolmogorov complexity via the Minimum Description Length (MDL) principle, using a score that is mini-max optimal with regard to the model class under consideration. This means that even in an adversarial setting, such as when the true distribution is not in this class, we still obtain the optimal encoding for the data relative to the class. We instantiate this framework, which we call CISC, for pairs of univariate discrete variables, using the class of multinomial distributions. Experiments show that CISC is highly accurate on synthetic, benchmark, as well as real-world data, outperforming the state of the art by a margin, and scales extremely well with regard to sample and domain sizes.

Title: Automated Speed and Lane Change Decision Making using Deep Reinforcement Learning, Abstract: This paper introduces a method, based on deep reinforcement learning, for automatically generating a general purpose decision making function. A Deep Q-Network agent was trained in a simulated environment to handle speed and lane change decisions for a truck-trailer combination. In a highway driving case, it is shown that the method produced an agent that matched or surpassed the performance of a commonly used reference model. To demonstrate the generality of the method, the exact same algorithm was also tested by training it for an overtaking case on a road with oncoming traffic. Furthermore, a novel way of applying a convolutional neural network to high level input that represents interchangeable objects is also introduced.

Title: Reducing biases on $H_0$ measurements using strong lensing and galaxy dynamics: results from the EAGLE simulation, Abstract: Cosmological parameter constraints from observations of time-delay lenses are becoming increasingly precise. However, there may be significant bias and scatter in these measurements due to, among other things, the so-called mass-sheet degeneracy. To estimate these uncertainties, we analyze strong lenses from the largest EAGLE hydrodynamical simulation. We apply a mass-sheet transformation to the radial density profiles of lenses, and by selecting lenses near isothermality, we find that the bias on H0 can be reduced to 5% with an intrinsic scatter of 10%, confirming previous results performed on a different simulation data set. We further investigate whether combining lensing observables with kinematic constraints helps to minimize this bias. We do not detect any significant dependence of the bias on lens model parameters or observational properties of the galaxy, but depending on the source--lens configuration, a bias may still exist. Cross lenses provide an accurate estimate of the Hubble constant, while fold (double) lenses tend to be biased low (high). With kinematic constraints, double lenses show bias and intrinsic scatter of 6% and 10%, respectively, while quad lenses show bias and intrinsic scatter of 0.5% and 10%, respectively. For lenses with a reduced $\chi^2 > 1$, a power-law dependence of the $\chi^2$ on the lens environment (number of nearby galaxies) is seen. Lastly, we model, in greater detail, the cases of two double lenses that are significantly biased. We are able to remove the bias, suggesting that the remaining biases could also be reduced by carefully taking into account additional sources of systematic uncertainty.

Title: Heisenberg Modules over Quantum 2-tori are metrized quantum vector bundles, Abstract: The modular Gromov-Hausdorff propinquity is a distance on classes of modules endowed with quantum metric information, in the form of a metric form of a connection and a left Hilbert module structure. This paper proves that the family of Heisenberg modules over quantum two tori, when endowed with their canonical connections, form a family of metrized quantum vector bundles, as a first step in proving that Heisenberg modules form a continuous family for the modular Gromov-Hausdorff propinquity.

Title: Toward Controlled Generation of Text, Abstract: Generic generation and manipulation of text is challenging and has limited success compared to recent deep generative modeling in visual domain. This paper aims at generating plausible natural language sentences, whose attributes are dynamically controlled by learning disentangled latent representations with designated semantics. We propose a new neural generative model which combines variational auto-encoders and holistic attribute discriminators for effective imposition of semantic structures. With differentiable approximation to discrete text samples, explicit constraints on independent attribute controls, and efficient collaborative learning of generator and discriminators, our model learns highly interpretable representations from even only word annotations, and produces realistic sentences with desired attributes. Quantitative evaluation validates the accuracy of sentence and attribute generation.

Title: Human peripheral blur is optimal for object recognition, Abstract: Our eyes sample a disproportionately large amount of information at the centre of gaze with increasingly sparse sampling into the periphery. This sampling scheme is widely believed to be a wiring constraint whereby high resolution at the centre is achieved by sacrificing spatial acuity in the periphery. Here we propose that this sampling scheme may be optimal for object recognition because the relevant spatial content is dense near an object and sparse in the surrounding vicinity. We tested this hypothesis by training deep convolutional neural networks on full-resolution and foveated images. Our main finding is that networks trained on images with foveated sampling show better object classification compared to networks trained on full resolution images. Importantly, blurring images according to the human blur function yielded the best performance compared to images with shallower or steeper blurring. Taken together our results suggest that, peripheral blurring in our eyes may have evolved for optimal object recognition, rather than merely to satisfy wiring constraints.

Title: Acceleration of Mean Square Distance Calculations with Floating Close Structure in Metadynamics Simulations, Abstract: Molecular dynamics simulates the~movements of atoms. Due to its high cost, many methods have been developed to "push the~simulation forward". One of them, metadynamics, can hasten the~molecular dynamics with the~help of variables describing the~simulated process. However, the~evaluation of these variables can include numerous mean square distance calculations that introduce substantial computational demands, thus jeopardize the~benefit of the~approach. Recently, we proposed an~approximative method that significantly reduces the~number of these distance calculations. Here we evaluate the~performance and the~scalability on two molecular systems. We assess the~maximal theoretical speed-up based on the reduction of distance computations and Ahmdal's law and compare it to the~practical speed-up achieved with our implementation.

Title: Neutronic Analysis on Potential Accident Tolerant Fuel-Cladding Combination U$_3$Si$_2$-FeCrAl, Abstract: Neutronic performance is investigated for a potential accident tolerant fuel (ATF),which consists of U$_3$Si$_2$ fuel and FeCrAl cladding. In comparison with current UO$_2$-Zr system, FeCrAl has a better oxidation resistance but a larger thermal neutron absorption cross section. U$_3$Si$_2$ has a higher thermal conductivity and a higher uranium density, which can compensate the reactivity suppressed by FeCrAl. Based on neutronic investigations, a possible U$_3$Si$_2$-FeCrAl fuel-cladding systemis taken into consideration. Fundamental properties of the suggested fuel-cladding combination are investigated in a fuel assembly.These properties include moderator and fuel temperature coefficients, control rods worth, radial power distribution (in a fuel rod), and different void reactivity coefficients. The present work proves that the new combination has less reactivity variation during its service lifetime. Although, compared with the current system, it has a little larger deviation on power distribution and a little less negative temperature coefficient and void reactivity coefficient and its control rods worth is less important, variations of these parameters are less important during the service lifetime of fuel. Hence, U$_3$Si$_2$-FeCrAl system is a potential ATF candidate from a neutronic view.

Title: A Macdonald refined topological vertex, Abstract: We consider the refined topological vertex of Iqbal et al, as a function of two parameters (x, y), and deform it by introducing Macdonald parameters (q, t), as in the work of Vuletic on plane partitions, to obtain 'a Macdonald refined topological vertex'. In the limit q -> t, we recover the refined topological vertex of Iqbal et al. In the limit x -> y, we obtain a qt-deformation of the topological vertex of Aganagic et al. Copies of the vertex can be glued to obtain qt-deformed 5D instanton partition functions that have well-defined 4D limits and, for generic values of (q, t), contain infinite-towers of poles for every pole in the limit q -> t.

Title: Estimating Achievable Range of Ground Robots Operating on Single Battery Discharge for Operational Efficacy Amelioration, Abstract: Mobile robots are increasingly being used to assist with active pursuit and law enforcement. One major limitation for such missions is the resource (battery) allocated to the robot. Factors like nature and agility of evader, terrain over which pursuit is being carried out, plausible traversal velocity and the amount of necessary data to be collected all influence how long the robot can last in the field and how far it can travel. In this paper, we develop an analytical model that analyzes the energy utilization for a variety of components mounted on a robot to estimate the maximum operational range achievable by the robot operating on a single battery discharge. We categorize the major consumers of energy as: 1.) ancillary robotic functions such as computation, communication, sensing etc., and 2.) maneuvering which involves propulsion, steering etc. Both these consumers draw power from the common power source but the achievable range is largely affected by the proportion of power available for maneuvering. For this case study, we performed experiments with real robots on planar and graded surfaces and evaluated the estimation error for each case.

Title: Likelihood ratio test for variance components in nonlinear mixed effects models, Abstract: Mixed effects models are widely used to describe heterogeneity in a population. A crucial issue when adjusting such a model to data consists in identifying fixed and random effects. From a statistical point of view, it remains to test the nullity of the variances of a given subset of random effects. Some authors have proposed to use the likelihood ratio test and have established its asymptotic distribution in some particular cases. Nevertheless, to the best of our knowledge, no general variance components testing procedure has been fully investigated yet. In this paper, we study the likelihood ratio test properties to test that the variances of a general subset of the random effects are equal to zero in both linear and nonlinear mixed effects model, extending the existing results. We prove that the asymptotic distribution of the test is a chi-bar-square distribution, that is to say a mixture of chi-square distributions, and we identify the corresponding weights. We highlight in particular that the limiting distribution depends on the presence of correlations between the random effects but not on the linear or nonlinear structure of the mixed effects model. We illustrate the finite sample size properties of the test procedure through simulation studies and apply the test procedure to two real datasets of dental growth and of coucal growth.

Title: Solving $\ell^p\!$-norm regularization with tensor kernels, Abstract: In this paper, we discuss how a suitable family of tensor kernels can be used to efficiently solve nonparametric extensions of $\ell^p$ regularized learning methods. Our main contribution is proposing a fast dual algorithm, and showing that it allows to solve the problem efficiently. Our results contrast recent findings suggesting kernel methods cannot be extended beyond Hilbert setting. Numerical experiments confirm the effectiveness of the method.

Title: Complementary views on electron spectra: From Fluctuation Diagnostics to real space correlations, Abstract: We study the relation between the microscopic properties of a many-body system and the electron spectra, experimentally accessible by photoemission. In a recent paper [Phys. Rev. Lett. 114, 236402 (2015)], we introduced the "fluctuation diagnostics" approach, to extract the dominant wave vector dependent bosonic fluctuations from the electronic self-energy. Here, we first reformulate the theory in terms of fermionic modes, to render its connection with resonance valence bond (RVB) fluctuations more transparent. Secondly, by using a large-U expansion, where U is the Coulomb interaction, we relate the fluctuations to real space correlations. Therefore, it becomes possible to study how electron spectra are related to charge, spin, superconductivity and RVB-like real space correlations, broadening the analysis of an earlier work [Phys. Rev. B 89, 245130 (2014)]. This formalism is applied to the pseudogap physics of the two-dimensional Hubbard model, studied in the dynamical cluster approximation. We perform calculations for embedded clusters with up to 32 sites, having three inequivalent K-points at the Fermi surface. We find that as U is increased, correlation functions gradually attain values consistent with an RVB state. This first happens for correlation functions involving the antinodal point and gradually spreads to the nodal point along the Fermi surface. Simultaneously a pseudogap opens up along the Fermi surface. We relate this to a crossover from a Kondo-like state to an RVB-like localized cluster state and to the presence of RVB and spin fluctuations. These changes are caused by a strong momentum dependence in the cluster bath-couplings along the Fermi surface. We also show, from a more algorithmic perspective, how the time-consuming calculations in fluctuation diagnostics can be drastically simplified.

Title: Extracting Automata from Recurrent Neural Networks Using Queries and Counterexamples, Abstract: We present a novel algorithm that uses exact learning and abstraction to extract a deterministic finite automaton describing the state dynamics of a given trained RNN. We do this using Angluin's L* algorithm as a learner and the trained RNN as an oracle. Our technique efficiently extracts accurate automata from trained RNNs, even when the state vectors are large and require fine differentiation.

Title: GBDT of discrete skew-selfadjoint Dirac systems and explicit solutions of the corresponding non-stationary problems, Abstract: Generalized Bäcklund-Darboux transformations (GBDTs) of discrete skew-selfadjoint Dirac systems have been successfully used for explicit solving of direct and inverse problems of Weyl-Titchmarsh theory. During explicit solving of the direct and inverse problems, we considered GBDTs of the trivial initial systems. However, GBDTs of arbitrary discrete skew-selfadjoint Dirac systems are important as well and we introduce these transformations in the present paper. The obtained results are applied to the construction of explicit solutions of the interesting related non-stationary systems.

Title: Smooth Neighbors on Teacher Graphs for Semi-supervised Learning, Abstract: The recently proposed self-ensembling methods have achieved promising results in deep semi-supervised learning, which penalize inconsistent predictions of unlabeled data under different perturbations. However, they only consider adding perturbations to each single data point, while ignoring the connections between data samples. In this paper, we propose a novel method, called Smooth Neighbors on Teacher Graphs (SNTG). In SNTG, a graph is constructed based on the predictions of the teacher model, i.e., the implicit self-ensemble of models. Then the graph serves as a similarity measure with respect to which the representations of "similar" neighboring points are learned to be smooth on the low-dimensional manifold. We achieve state-of-the-art results on semi-supervised learning benchmarks. The error rates are 9.89%, 3.99% for CIFAR-10 with 4000 labels, SVHN with 500 labels, respectively. In particular, the improvements are significant when the labels are fewer. For the non-augmented MNIST with only 20 labels, the error rate is reduced from previous 4.81% to 1.36%. Our method also shows robustness to noisy labels.

Title: Achieving Spectrum Efficient Communication Under Cross-Technology Interference, Abstract: In wireless communication, heterogeneous technologies such as WiFi, ZigBee and BlueTooth operate in the same ISM band.With the exponential growth in the number of wireless devices, the ISM band becomes more and more crowded. These heterogeneous devices have to compete with each other to access spectrum resources, generating cross-technology interference (CTI). Since CTI may destroy wireless communication, this field is facing an urgent and challenging need to investigate spectrum efficiency under CTI. In this paper, we introduce a novel framework to address this problem from two aspects. On the one hand, from the perspective of each communication technology itself, we propose novel channel/link models to capture the channel/link status under CTI. On the other hand, we investigate spectrum efficiency from the perspective by taking all heterogeneous technologies as a whole and building crosstechnology communication among them. The capability of direct communication among heterogeneous devices brings great opportunities to harmoniously sharing the spectrum with collaboration rather than competition.

Title: A Galactic Cosmic Ray Electron Intensity Increase of a factor of up to 100 At Energies between 3 and 50 MeV in the Heliosheath between the Termination Shock and the Heliopause Due to Solar Modulation As Measured by Voyager 1, Abstract: We have derived background corrected intensities of 3-50 MeV galactic electrons observed by Voyager 1 as it passes through the heliosheath from 95 to 122 AU. The overall intensity change of the background corrected data from the inner to the outer boundary of the heliosheath is a maximum of a factor ~100 at 15 MeV. At lower energies this fractional change becomes less and the corrected electron spectra in the heliosheath becomes progressively steeper, reaching values ~ -2.5 for the spectral index just outside of the termination shock. At higher energies the spectra of electrons has an exponent changing from the negative LIS spectral index of -1.3 to values approaching zero in the heliosheath as a result of the solar modulation of the galactic electron component. The large modulation effects observed below ~100 MV are possible evidence for enhanced diffusion as part of the modulation process for electrons in the heliosheath.

Title: On self-affine sets, Abstract: We survey the dimension theory of self-affine sets for general mathematical audience. The article is in Finnish.

Title: Multiplicative slices, relativistic Toda and shifted quantum affine algebras, Abstract: We introduce the shifted quantum affine algebras. They map homomorphically into the quantized $K$-theoretic Coulomb branches of $3d\ {\mathcal N}=4$ SUSY quiver gauge theories. In type $A$, they are endowed with a coproduct, and they act on the equivariant $K$-theory of parabolic Laumon spaces. In type $A_1$, they are closely related to the open relativistic quantum Toda lattice of type $A$.

Title: High temperature thermodynamics of the honeycomb-lattice Kitaev-Heisenberg model: A high temperature series expansion study, Abstract: We develop high temperature series expansions for the thermodynamic properties of the honeycomb-lattice Kitaev-Heisenberg model. Numerical results for uniform susceptibility, heat capacity and entropy as a function of temperature for different values of the Kitaev coupling $K$ and Heisenberg exachange coupling $J$ (with $|J|\le |K|$) are presented. These expansions show good convergence down to a temperature of a fraction of $K$ and in some cases down to $T=K/10$. In the Kitaev exchange dominated regime, the inverse susceptibility has a nearly linear temperature dependence over a wide temperature range. However, we show that already at temperatures $10$-times the Curie-Weiss temperature, the effective Curie-Weiss constant estimated from the data can be off by a factor of 2. We find that the magnitude of the heat capacity maximum at the short-range order peak, is substantially smaller for small $J/K$ than for $J$ of order or larger than $K$. We suggest that this itself represents a simple marker for the relative importance of the Kitaev terms in these systems. Somewhat surprisingly, both heat capacity and susceptibility data on Na$_2$IrO$_3$ are consistent with a dominant {\it antiferromagnetic} Kitaev exchange constant of about $300-400$ $K$.

Title: Inference in Sparse Graphs with Pairwise Measurements and Side Information, Abstract: We consider the statistical problem of recovering a hidden "ground truth" binary labeling for the vertices of a graph up to low Hamming error from noisy edge and vertex measurements. We present new algorithms and a sharp finite-sample analysis for this problem on trees and sparse graphs with poor expansion properties such as hypergrids and ring lattices. Our method generalizes and improves over that of Globerson et al. (2015), who introduced the problem for two-dimensional grid lattices. For trees we provide a simple, efficient, algorithm that infers the ground truth with optimal Hamming error has optimal sample complexity and implies recovery results for all connected graphs. Here, the presence of side information is critical to obtain a non-trivial recovery rate. We then show how to adapt this algorithm to tree decompositions of edge-subgraphs of certain graph families such as lattices, resulting in optimal recovery error rates that can be obtained efficiently The thrust of our analysis is to 1) use the tree decomposition along with edge measurements to produce a small class of viable vertex labelings and 2) apply an analysis influenced by statistical learning theory to show that we can infer the ground truth from this class using vertex measurements. We show the power of our method in several examples including hypergrids, ring lattices, and the Newman-Watts model for small world graphs. For two-dimensional grids, our results improve over Globerson et al. (2015) by obtaining optimal recovery in the constant-height regime.

Title: Guiding Reinforcement Learning Exploration Using Natural Language, Abstract: In this work we present a technique to use natural language to help reinforcement learning generalize to unseen environments. This technique uses neural machine translation, specifically the use of encoder-decoder networks, to learn associations between natural language behavior descriptions and state-action information. We then use this learned model to guide agent exploration using a modified version of policy shaping to make it more effective at learning in unseen environments. We evaluate this technique using the popular arcade game, Frogger, under ideal and non-ideal conditions. This evaluation shows that our modified policy shaping algorithm improves over a Q-learning agent as well as a baseline version of policy shaping.

Title: Cell Coverage Extension with Orthogonal Random Precoding for Massive MIMO Systems, Abstract: In this paper, we investigate a coverage extension scheme based on orthogonal random precoding (ORP) for the downlink of massive multiple-input multiple-output (MIMO) systems. In this scheme, a precoding matrix consisting of orthogonal vectors is employed at the transmitter to enhance the maximum signal-to-interference-plus-noise ratio (SINR) of the user. To analyze and optimize the ORP scheme in terms of cell coverage, we derive the analytical expressions of the downlink coverage probability for two receiver structures, namely, the single-antenna (SA) receiver and multiple-antenna receiver with antenna selection (AS). The simulation results show that the analytical expressions accurately capture the coverage behaviors of the systems employing the ORP scheme. It is also shown that the optimal coverage performance is achieved when a single precoding vector is used under the condition that the threshold of the signal-to-noise ratio of the coverage is greater than one. The performance of the ORP scheme is further analyzed when different random precoder groups are utilized over multiple time slots to exploit precoding diversity. The numerical results show that the proposed ORP scheme over multiple time slots provides a substantial coverage gain over the space-time coding scheme despite its low feedback overhead.

Title: Quasiparticles and charge transfer at the two surfaces of the honeycomb iridate Na$_2$IrO$_3$, Abstract: Direct experimental investigations of the low-energy electronic structure of the Na$_2$IrO$_3$ iridate insulator are sparse and draw two conflicting pictures. One relies on flat bands and a clear gap, the other involves dispersive states approaching the Fermi level, pointing to surface metallicity. Here, by a combination of angle-resolved photoemission, photoemission electron microscopy, and x-ray absorption, we show that the correct picture is more complex and involves an anomalous band, arising from charge transfer from Na atoms to Ir-derived states. Bulk quasiparticles do exist, but in one of the two possible surface terminations the charge transfer is smaller and they remain elusive.

Title: Breaking Bivariate Records, Abstract: We establish a fundamental property of bivariate Pareto records for independent observations uniformly distributed in the unit square. We prove that the asymptotic conditional distribution of the number of records broken by an observation given that the observation sets a record is Geometric with parameter 1/2.

Title: A Bag-of-Paths Node Criticality Measure, Abstract: This work compares several node (and network) criticality measures quantifying to which extend each node is critical with respect to the communication flow between nodes of the network, and introduces a new measure based on the Bag-of-Paths (BoP) framework. Network disconnection simulation experiments show that the new BoP measure outperforms all the other measures on a sample of Erdos-Renyi and Albert-Barabasi graphs. Furthermore, a faster (still O(n^3)), approximate, BoP criticality relying on the Sherman-Morrison rank-one update of a matrix is introduced for tackling larger networks. This approximate measure shows similar performances as the original, exact, one.

Title: Selection of quasi-stationary states in the Navier-Stokes equation on the torus, Abstract: The two dimensional incompressible Navier-Stokes equation on $D_\delta := [0, 2\pi\delta] \times [0, 2\pi]$ with $\delta \approx 1$, periodic boundary conditions, and viscosity $0 < \nu \ll 1$ is considered. Bars and dipoles, two explicitly given quasi-stationary states of the system, evolve on the time scale $\mathcal{O}(e^{-\nu t})$ and have been shown to play a key role in its long-time evolution. Of particular interest is the role that $\delta$ plays in selecting which of these two states is observed. Recent numerical studies suggest that, after a transient period of rapid decay of the high Fourier modes, the bar state will be selected if $\delta \neq 1$, while the dipole will be selected if $\delta = 1$. Our results support this claim and seek to mathematically formalize it. We consider the system in Fourier space, project it onto a center manifold consisting of the lowest eight Fourier modes, and use this as a model to study the selection of bars and dipoles. It is shown for this ODE model that the value of $\delta$ controls the behavior of the asymptotic ratio of the low modes, thus determining the likelihood of observing a bar state or dipole after an initial transient period. Moreover, in our model, for all $\delta \approx 1$, there is an initial time period in which the high modes decay at the rapid rate $\mathcal{O}(e^{-t/\nu})$, while the low modes evolve at the slower $\mathcal{O}(e^{-\nu t})$ rate. The results for the ODE model are proven using energy estimates and invariant manifolds and further supported by formal asymptotic expansions and numerics.

Title: Geometric Enclosing Networks, Abstract: Training model to generate data has increasingly attracted research attention and become important in modern world applications. We propose in this paper a new geometry-based optimization approach to address this problem. Orthogonal to current state-of-the-art density-based approaches, most notably VAE and GAN, we present a fresh new idea that borrows the principle of minimal enclosing ball to train a generator G\left(\bz\right) in such a way that both training and generated data, after being mapped to the feature space, are enclosed in the same sphere. We develop theory to guarantee that the mapping is bijective so that its inverse from feature space to data space results in expressive nonlinear contours to describe the data manifold, hence ensuring data generated are also lying on the data manifold learned from training data. Our model enjoys a nice geometric interpretation, hence termed Geometric Enclosing Networks (GEN), and possesses some key advantages over its rivals, namely simple and easy-to-control optimization formulation, avoidance of mode collapsing and efficiently learn data manifold representation in a completely unsupervised manner. We conducted extensive experiments on synthesis and real-world datasets to illustrate the behaviors, strength and weakness of our proposed GEN, in particular its ability to handle multi-modal data and quality of generated data.

Title: Out-degree reducing partitions of digraphs, Abstract: Let $k$ be a fixed integer. We determine the complexity of finding a $p$-partition $(V_1, \dots, V_p)$ of the vertex set of a given digraph such that the maximum out-degree of each of the digraphs induced by $V_i$, ($1\leq i\leq p$) is at least $k$ smaller than the maximum out-degree of $D$. We show that this problem is polynomial-time solvable when $p\geq 2k$ and ${\cal NP}$-complete otherwise. The result for $k=1$ and $p=2$ answers a question posed in \cite{bangTCS636}. We also determine, for all fixed non-negative integers $k_1,k_2,p$, the complexity of deciding whether a given digraph of maximum out-degree $p$ has a $2$-partition $(V_1,V_2)$ such that the digraph induced by $V_i$ has maximum out-degree at most $k_i$ for $i\in [2]$. It follows from this characterization that the problem of deciding whether a digraph has a 2-partition $(V_1,V_2)$ such that each vertex $v\in V_i$ has at least as many neighbours in the set $V_{3-i}$ as in $V_i$, for $i=1,2$ is ${\cal NP}$-complete. This solves a problem from \cite{kreutzerEJC24} on majority colourings.

Title: Presymplectic convexity and (ir)rational polytopes, Abstract: In this paper, we extend the Atiyah--Guillemin--Sternberg convexity theorem and Delzant's classification of symplectic toric manifolds to presymplectic manifolds. We also define and study the Morita equivalence of presymplectic toric manifolds and of their corresponding framed momentum polytopes, which may be rational or non-rational. Toric orbifolds, quasifolds and non-commutative toric varieties may be viewed as the quotient of our presymplectic toric manifolds by the kernel isotropy foliation of the presymplectic form.

Title: Unsupervised Learning of Mixture Regression Models for Longitudinal Data, Abstract: This paper is concerned with learning of mixture regression models for individuals that are measured repeatedly. The adjective "unsupervised" implies that the number of mixing components is unknown and has to be determined, ideally by data driven tools. For this purpose, a novel penalized method is proposed to simultaneously select the number of mixing components and to estimate the mixing proportions and unknown parameters in the models. The proposed method is capable of handling both continuous and discrete responses by only requiring the first two moment conditions of the model distribution. It is shown to be consistent in both selecting the number of components and estimating the mixing proportions and unknown regression parameters. Further, a modified EM algorithm is developed to seamlessly integrate model selection and estimation. Simulation studies are conducted to evaluate the finite sample performance of the proposed procedure. And it is further illustrated via an analysis of a primary biliary cirrhosis data set.

Title: Anomalous electron states, Abstract: By the certain macroscopic perturbations in condensed matter anomalous electron wells can be formed due to a local reduction of electromagnetic zero point energy. These wells are narrow, of the width $\sim 10^{-11}cm$, and with the depth $\sim 1MeV$. Such anomalous states, from the formal standpoint of quantum mechanics, correspond to a singular solution of a wave equation produced by the non-physical $\delta(\vec R)$ source. The resolution, on the level of the Standard Model, of the tiny region around the formal singularity shows that the state is physical. The creation of those states in an atomic system is of the formal probability $\exp(-1000)$. The probability becomes not small under a perturbation which rapidly varies in space, on the scale $10^{-11}cm$. In condensed matter such perturbation may relate to acoustic shock waves. In this process the short scale is the length of the standing de Broglie wave of a reflected lattice atom. Under electron transitions in the anomalous well (anomalous atom) $keV$ X-rays are expected to be emitted. A macroscopic amount of anomalous atoms, of the size $10^{-11}cm$ each, can be formed in a solid resulting in ${\it collapsed}$ ${\it matter}$ with $10^9$ times enhanced density.

Title: Temporal processing and context dependency in C. elegans mechanosensation, Abstract: A quantitative understanding of how sensory signals are transformed into motor outputs places useful constraints on brain function and helps reveal the brain's underlying computations. We investigate how the nematode C. elegans responds to time-varying mechanosensory signals using a high-throughput optogenetic assay and automated behavior quantification. In the prevailing picture of the touch circuit, the animal's behavior is determined by which neurons are stimulated and by the stimulus amplitude. In contrast, we find that the behavioral response is tuned to temporal properties of mechanosensory signals, like its integral and derivative, that extend over many seconds. Mechanosensory signals, even in the same neurons, can be tailored to elicit different behavioral responses. Moreover, we find that the animal's response also depends on its behavioral context. Most dramatically, the animal ignores all tested mechanosensory stimuli during turns. Finally, we present a linear-nonlinear model that predicts the animal's behavioral response to stimulus.

Title: On the putative essential discreteness of q-generalized entropies, Abstract: It has been argued in [EPL {\bf 90} (2010) 50004], entitled {\it Essential discreteness in generalized thermostatistics with non-logarithmic entropy}, that "continuous Hamiltonian systems with long-range interactions and the so-called q-Gaussian momentum distributions are seen to be outside the scope of non-extensive statistical mechanics". The arguments are clever and appealing. We show here that, however, some mathematical subtleties render them unconvincing

Title: Emergent electronic structure of CaFe2As2, Abstract: CaFe2As2 exhibits collapsed tetragonal (cT) structure and varied exotic behavior under pressure at low temperatures that led to debate on linking the structural changes to its exceptional electronic properties like superconductivity, magnetism, etc. Here, we investigate the electronic structure of CaFe2As2 forming in different structures employing density functional theory. The results indicate better stability of the cT phase with enhancement in hybridization induced effects and shift of the energy bands towards lower energies. The Fermi surface centered around $\Gamma$ point gradually vanishes with the increase in pressure. Consequently, the nesting between the hole and electron Fermi surfaces associated to the spin density wave state disappears indicating a pathway to achieve the proximity to quantum fluctuations. The magnetic moment at the Fe sites diminishes in the cT phase consistent with the magnetic susceptibility results. Notably, the hybridization of Ca 4s states (Ca-layer may be treated as a charge reservoir layer akin to those in cuprate superconductors) is significantly enhanced in the cT phase revealing its relevance in its interesting electronic properties.

Title: Smart "Predict, then Optimize", Abstract: Many real-world analytics problems involve two significant challenges: prediction and optimization. Due to the typically complex nature of each challenge, the standard paradigm is to predict, then optimize. By and large, machine learning tools are intended to minimize prediction error and do not account for how the predictions will be used in a downstream optimization problem. In contrast, we propose a new and very general framework, called Smart "Predict, then Optimize" (SPO), which directly leverages the optimization problem structure, i.e., its objective and constraints, for designing successful analytics tools. A key component of our framework is the SPO loss function, which measures the quality of a prediction by comparing the objective values of the solutions generated using the predicted and observed parameters, respectively. Training a model with respect to the SPO loss is computationally challenging, and therefore we also develop a surrogate loss function, called the SPO+ loss, which upper bounds the SPO loss, has desirable convexity properties, and is statistically consistent under mild conditions. We also propose a stochastic gradient descent algorithm which allows for situations in which the number of training samples is large, model regularization is desired, and/or the optimization problem of interest is nonlinear or integer. Finally, we perform computational experiments to empirically verify the success of our SPO framework in comparison to the standard predict-then-optimize approach.

Title: Density estimation on small datasets, Abstract: How might a smooth probability distribution be estimated, with accurately quantified uncertainty, from a limited amount of sampled data? Here we describe a field-theoretic approach that addresses this problem remarkably well in one dimension, providing an exact nonparametric Bayesian posterior without relying on tunable parameters or large-data approximations. Strong non-Gaussian constraints, which require a non-perturbative treatment, are found to play a major role in reducing distribution uncertainty. A software implementation of this method is provided.

Title: Security Trust Zone in 5G Networks, Abstract: Fifth Generation (5G) telecommunication system is going to deliver a flexible radio access network (RAN). Security functions such as authorization, authentication and accounting (AAA) are expected to be distributed from central clouds to edge clouds. We propose a novel architectural security solution that applies to 5G networks. It is called Trust Zone (TZ) that is designed as an enhancement of the 5G AAA in the edge cloud. TZ also provides an autonomous and decentralized security policy for different tenants under variable network conditions. TZ also initiates an ability of disaster cognition and extends the security functionalities to a set of flexible and highly available emergency services in the edge cloud.

Title: Few-shot learning of neural networks from scratch by pseudo example optimization, Abstract: In this paper, we propose a simple but effective method for training neural networks with a limited amount of training data. Our approach inherits the idea of knowledge distillation that transfers knowledge from a deep or wide reference model to a shallow or narrow target model. The proposed method employs this idea to mimic predictions of reference estimators that are more robust against overfitting than the network we want to train. Different from almost all the previous work for knowledge distillation that requires a large amount of labeled training data, the proposed method requires only a small amount of training data. Instead, we introduce pseudo training examples that are optimized as a part of model parameters. Experimental results for several benchmark datasets demonstrate that the proposed method outperformed all the other baselines, such as naive training of the target model and standard knowledge distillation.

Title: Identities and congruences involving the Fubini polynomials, Abstract: In this paper, we investigate the umbral representation of the Fubini polynomials $F_{x}^{n}:=F_{n}(x)$ to derive some properties involving these polynomials. For any prime number $p$ and any polynomial $f$ with integer coefficients, we show $(f(F_{x}))^{p}\equiv f(F_{x})$ and we give other curious congruences.

Title: High-sensitivity Kinetic Inductance Detectors for CALDER, Abstract: Providing a background discrimination tool is crucial for enhancing the sensitivity of next-generation experiments searching for neutrinoless double- beta decay. The development of high-sensitivity (< 20 eV RMS) cryogenic light detectors allows simultaneous read-out of the light and heat signals and enables background suppression through particle identification. The Cryogenic wide- Area Light Detector with Excellent Resolution (CALDER) R&D already proved the potential of this technique using the phonon-mediated Kinetic Inductance Detectors (KIDs) approach. The first array prototype with 4 Aluminum KIDs on a 2 $\times$ 2 cm2 Silicon substrate showed a baseline resolution of 154 $\pm$ 7 eV RMS. Improving the design and the readout of the resonator, the next CALDER prototype featured an energy resolution of 82 $\pm$ 4 eV, by sampling the same substrate with a single Aluminum KID.

Title: An Exploration of Approaches to Integrating Neural Reranking Models in Multi-Stage Ranking Architectures, Abstract: We explore different approaches to integrating a simple convolutional neural network (CNN) with the Lucene search engine in a multi-stage ranking architecture. Our models are trained using the PyTorch deep learning toolkit, which is implemented in C/C++ with a Python frontend. One obvious integration strategy is to expose the neural network directly as a service. For this, we use Apache Thrift, a software framework for building scalable cross-language services. In exploring alternative architectures, we observe that once trained, the feedforward evaluation of neural networks is quite straightforward. Therefore, we can extract the parameters of a trained CNN from PyTorch and import the model into Java, taking advantage of the Java Deeplearning4J library for feedforward evaluation. This has the advantage that the entire end-to-end system can be implemented in Java. As a third approach, we can extract the neural network from PyTorch and "compile" it into a C++ program that exposes a Thrift service. We evaluate these alternatives in terms of performance (latency and throughput) as well as ease of integration. Experiments show that feedforward evaluation of the convolutional neural network is significantly slower in Java, while the performance of the compiled C++ network does not consistently beat the PyTorch implementation.

Title: Monotonicity patterns and functional inequalities for classical and generalized Wright functions, Abstract: In this paper our aim is to present the completely monotonicity and convexity properties for the Wright function. As consequences of these results, we present some functional inequalities. Moreover, we derive the monotonicity and log-convexity results for the generalized Wright functions. As applications, we present several new inequalities (like Turán type inequalities) and we prove some geometric properties for four--parametric Mittag--Leffler functions.

Title: Galaxy And Mass Assembly: the evolution of the cosmic spectral energy distribution from z = 1 to z = 0, Abstract: We present the evolution of the Cosmic Spectral Energy Distribution (CSED) from $z = 1 - 0$. Our CSEDs originate from stacking individual spectral energy distribution fits based on panchromatic photometry from the Galaxy and Mass Assembly (GAMA) and COSMOS datasets in ten redshift intervals with completeness corrections applied. Below $z = 0.45$, we have credible SED fits from 100 nm to 1 mm. Due to the relatively low sensitivity of the far-infrared data, our far-infrared CSEDs contain a mix of predicted and measured fluxes above $z = 0.45$. Our results include appropriate errors to highlight the impact of these corrections. We show that the bolometric energy output of the Universe has declined by a factor of roughly four -- from $5.1 \pm 1.0$ at $z \sim 1$ to $1.3 \pm 0.3 \times 10^{35}~h_{70}$~W~Mpc$^{-3}$ at the current epoch. We show that this decrease is robust to cosmic variance, SED modelling and other various types of error. Our CSEDs are also consistent with an increase in the mean age of stellar populations. We also show that dust attenuation has decreased over the same period, with the photon escape fraction at 150~nm increasing from $16 \pm 3$ at $z \sim 1$ to $24 \pm 5$ per cent at the current epoch, equivalent to a decrease in $A_\mathrm{FUV}$ of 0.4~mag. Our CSEDs account for $68 \pm 12$ and $61 \pm 13$ per cent of the cosmic optical and infrared backgrounds respectively as defined from integrated galaxy counts and are consistent with previous estimates of the cosmic infrared background with redshift.

Title: Asymptotic formula of the number of Newton polygons, Abstract: In this paper, we enumerate Newton polygons asymptotically. The number of Newton polygons is computable by a simple recurrence equation, but unexpectedly the asymptotic formula of its logarithm contains growing oscillatory terms. As the terms come from non-trivial zeros of the Riemann zeta function, an estimation of the amplitude of the oscillating part is equivalent to the Riemann hypothesis.

Title: Stochastic Variance Reduction Methods for Policy Evaluation, Abstract: Policy evaluation is a crucial step in many reinforcement-learning procedures, which estimates a value function that predicts states' long-term value under a given policy. In this paper, we focus on policy evaluation with linear function approximation over a fixed dataset. We first transform the empirical policy evaluation problem into a (quadratic) convex-concave saddle point problem, and then present a primal-dual batch gradient method, as well as two stochastic variance reduction methods for solving the problem. These algorithms scale linearly in both sample size and feature dimension. Moreover, they achieve linear convergence even when the saddle-point problem has only strong concavity in the dual variables but no strong convexity in the primal variables. Numerical experiments on benchmark problems demonstrate the effectiveness of our methods.

Title: Measuring the polarization of electromagnetic fields using Rabi-rate measurements with spatial resolution: experiment and theory, Abstract: When internal states of atoms are manipulated using coherent optical or radio-frequency (RF) radiation, it is essential to know the polarization of the radiation with respect to the quantization axis of the atom. We first present a measurement of the two-dimensional spatial distribution of the electric-field amplitude of a linearly-polarized pulsed RF electric field at $\sim 25.6\,$GHz and its angle with respect to a static electric field. The measurements exploit coherent population transfer between the $35$s and $35$p Rydberg states of helium atoms in a pulsed supersonic beam. Based on this experimental result, we develop a general framework in the form of a set of equations relating the five independent polarization parameters of a coherently oscillating field in a fixed laboratory frame to Rabi rates of transitions between a ground and three excited states of an atom with arbitrary quantization axis. We then explain how these equations can be used to fully characterize the polarization in a minimum of five Rabi rate measurements by rotation of an external bias-field, or, knowing the polarization of the driving field, to determine the orientation of the static field using two measurements. The presented technique is not limited to Rydberg atoms and RF fields but can also be applied to characterize optical fields. The technique has potential for sensing the spatiotemporal properties of electromagnetic fields, e.g., in metrology devices or in hybrid experiments involving atoms close to surfaces.

Title: Software-based Microarchitectural Attacks, Abstract: Modern processors are highly optimized systems where every single cycle of computation time matters. Many optimizations depend on the data that is being processed. Software-based microarchitectural attacks exploit effects of these optimizations. Microarchitectural side-channel attacks leak secrets from cryptographic computations, from general purpose computations, or from the kernel. This leakage even persists across all common isolation boundaries, such as processes, containers, and virtual machines. Microarchitectural fault attacks exploit the physical imperfections of modern computer systems. Shrinking process technology introduces effects between isolated hardware elements that can be exploited by attackers to take control of the entire system. These attacks are especially interesting in scenarios where the attacker is unprivileged or even sandboxed. In this thesis, we focus on microarchitectural attacks and defenses on commodity systems. We investigate known and new side channels and show that microarchitectural attacks can be fully automated. Furthermore, we show that these attacks can be mounted in highly restricted environments such as sandboxed JavaScript code in websites. We show that microarchitectural attacks exist on any modern computer system, including mobile devices (e.g., smartphones), personal computers, and commercial cloud systems. This thesis consists of two parts. In the first part, we provide background on modern processor architectures and discuss state-of-the-art attacks and defenses in the area of microarchitectural side-channel attacks and microarchitectural fault attacks. In the second part, a selection of our papers are provided without modification from their original publications. I have co-authored these papers, which have subsequently been anonymously peer-reviewed, accepted, and presented at renowned international conferences.

Title: Boundary problems for the fractional and tempered fractional operators, Abstract: For characterizing the Brownian motion in a bounded domain: $\Omega$, it is well-known that the boundary conditions of the classical diffusion equation just rely on the given information of the solution along the boundary of a domain; on the contrary, for the Lévy flights or tempered Lévy flights in a bounded domain, it involves the information of a solution in the complementary set of $\Omega$, i.e., $\mathbb{R}^n\backslash \Omega$, with the potential reason that paths of the corresponding stochastic process are discontinuous. Guided by probability intuitions and the stochastic perspectives of anomalous diffusion, we show the reasonable ways, ensuring the clear physical meaning and well-posedness of the partial differential equations (PDEs), of specifying `boundary' conditions for space fractional PDEs modeling the anomalous diffusion. Some properties of the operators are discussed, and the well-posednesses of the PDEs with generalized boundary conditions are proved.

Title: Bayesian Unification of Gradient and Bandit-based Learning for Accelerated Global Optimisation, Abstract: Bandit based optimisation has a remarkable advantage over gradient based approaches due to their global perspective, which eliminates the danger of getting stuck at local optima. However, for continuous optimisation problems or problems with a large number of actions, bandit based approaches can be hindered by slow learning. Gradient based approaches, on the other hand, navigate quickly in high-dimensional continuous spaces through local optimisation, following the gradient in fine grained steps. Yet, apart from being susceptible to local optima, these schemes are less suited for online learning due to their reliance on extensive trial-and-error before the optimum can be identified. In this paper, we propose a Bayesian approach that unifies the above two paradigms in one single framework, with the aim of combining their advantages. At the heart of our approach we find a stochastic linear approximation of the function to be optimised, where both the gradient and values of the function are explicitly captured. This allows us to learn from both noisy function and gradient observations, and predict these properties across the action space to support optimisation. We further propose an accompanying bandit driven exploration scheme that uses Bayesian credible bounds to trade off exploration against exploitation. Our empirical results demonstrate that by unifying bandit and gradient based learning, one obtains consistently improved performance across a wide spectrum of problem environments. Furthermore, even when gradient feedback is unavailable, the flexibility of our model, including gradient prediction, still allows us outperform competing approaches, although with a smaller margin. Due to the pervasiveness of bandit based optimisation, our scheme opens up for improved performance both in meta-optimisation and in applications where gradient related information is readily available.

Title: Detecting Bot Activity in the Ethereum Blockchain Network, Abstract: The Ethereum blockchain network is a decentralized platform enabling smart contract execution and transactions of Ether (ETH) [1], its designated cryptocurrency. Ethereum is the second most popular cryptocurrency with a market cap of more than 100 billion USD, with hundreds of thousands of transactions executed daily by hundreds of thousands of unique wallets. Tens of thousands of those wallets are newly generated each day. The Ethereum platform enables anyone to freely open multiple new wallets [2] free of charge (resulting in a large number of wallets that are controlled by the same entities). This attribute makes the Ethereum network a breeding space for activity by software robots (bots). The existence of bots is widespread in different digital technologies and there are various approaches to detect their activity such as rule-base, clustering, machine learning and more [3,4]. In this work we demonstrate how bot detection can be implemented using a network theory approach.

Title: Near-infrared laser thermal conjunctivoplasty, Abstract: Conjunctivochalasis is a common cause of tear dysfunction due to the conjunctiva becoming loose and wrinkly with age. The current solutions to this disease include either surgical excision in the operating room, or thermoreduction of the loose tissue with hot wire in the clinic. We developed a near-infrared (NIR) laser thermal conjunctivoplasty (LTC) system, which gently shrinks the redundant tissue. The NIR light is mainly absorbed by water, so the heating is even and there is no bleeding. The system utilizes a 1460-nm programmable laser diode system as a light source. A miniaturized handheld probe delivers the laser light and focuses the laser into a 10x1 mm2 line. A foot pedal is used to deliver a preset number of calibrated laser pulses. A fold of loose conjunctiva is grasped by a pair of forceps. The infrared laser light is delivered through an optical fiber and a laser line is focused exactly on the conjunctival fold by a cylindrical lens. Ex vivo experiments using porcine eye were performed with the optimal laser parameters. It was found that up to 50% of conjunctiva shrinkage could be achieved.

Title: VTA: An Open Hardware-Software Stack for Deep Learning, Abstract: Hardware acceleration is an enabler for ubiquitous and efficient deep learning. With hardware accelerators being introduced in datacenter and edge devices, it is time to acknowledge that hardware specialization is central to the deep learning system stack. This technical report presents the Versatile Tensor Accelerator (VTA), an open, generic, and customizable deep learning accelerator design. VTA is a programmable accelerator that exposes a RISC-like programming abstraction to describe operations at the tensor level. We designed VTA to expose the most salient and common characteristics of mainstream deep learning accelerators, such as tensor operations, DMA load/stores, and explicit compute/memory arbitration. VTA is more than a standalone accelerator design: it's an end-to-end solution that includes drivers, a JIT runtime, and an optimizing compiler stack based on TVM. The current release of VTA includes a behavioral hardware simulator, as well as the infrastructure to deploy VTA on low-cost FPGA development boards for fast prototyping. By extending the TVM stack with a customizable, and open source deep learning hardware accelerator design, we are exposing a transparent end-to-end deep learning stack from the high-level deep learning framework, down to the actual hardware design and implementation. This forms a truly end-to-end, from software-to-hardware open source stack for deep learning systems.

Title: Exact partial information decompositions for Gaussian systems based on dependency constraints, Abstract: The Partial Information Decomposition (PID) [arXiv:1004.2515] provides a theoretical framework to characterize and quantify the structure of multivariate information sharing. A new method (Idep) has recently been proposed for computing a two-predictor PID over discrete spaces. [arXiv:1709.06653] A lattice of maximum entropy probability models is constructed based on marginal dependency constraints, and the unique information that a particular predictor has about the target is defined as the minimum increase in joint predictor-target mutual information when that particular predictor-target marginal dependency is constrained. Here, we apply the Idep approach to Gaussian systems, for which the marginally constrained maximum entropy models are Gaussian graphical models. Closed form solutions for the Idep PID are derived for both univariate and multivariate Gaussian systems. Numerical and graphical illustrations are provided, together with practical and theoretical comparisons of the Idep PID with the minimum mutual information PID (Immi). [arXiv:1411.2832] In particular, it is proved that the Immi method generally produces larger estimates of redundancy and synergy than does the Idep method. In discussion of the practical examples, the PIDs are complemented by the use of deviance tests for the comparison of Gaussian graphical models.

Title: Finite-time generalization of the thermodynamic uncertainty relation, Abstract: For fluctuating currents in non-equilibrium steady states, the recently discovered thermodynamic uncertainty relation expresses a fundamental relation between their variance and the overall entropic cost associated with the driving. We show that this relation holds not only for the long-time limit of fluctuations, as described by large deviation theory, but also for fluctuations on arbitrary finite time scales. This generalization facilitates applying the thermodynamic uncertainty relation to single molecule experiments, for which infinite timescales are not accessible. Importantly, often this finite-time variant of the relation allows inferring a bound on the entropy production that is even stronger than the one obtained from the long-time limit. We illustrate the relation for the fluctuating work that is performed by a stochastically switching laser tweezer on a trapped colloidal particle.

Title: Composition Properties of Inferential Privacy for Time-Series Data, Abstract: With the proliferation of mobile devices and the internet of things, developing principled solutions for privacy in time series applications has become increasingly important. While differential privacy is the gold standard for database privacy, many time series applications require a different kind of guarantee, and a number of recent works have used some form of inferential privacy to address these situations. However, a major barrier to using inferential privacy in practice is its lack of graceful composition -- even if the same or related sensitive data is used in multiple releases that are safe individually, the combined release may have poor privacy properties. In this paper, we study composition properties of a form of inferential privacy called Pufferfish when applied to time-series data. We show that while general Pufferfish mechanisms may not compose gracefully, a specific Pufferfish mechanism, called the Markov Quilt Mechanism, which was recently introduced, has strong composition properties comparable to that of pure differential privacy when applied to time series data.

Title: Well-posedness of a Model for the Growth of Tree Stems and Vines, Abstract: The paper studies a PDE model for the growth of a tree stem or a vine, having the form of a differential inclusion with state constraints. The equations describe the elongation due to cell growth, and the response to gravity and to external obstacles. The main theorem shows that the evolution problem is well posed, until a specific "breakdown configuration" is reached. A formula is proved, characterizing the reaction produced by unilateral constraints. At a.e. time t, this is determined by the minimization of an elastic energy functional under suitable constraints.

Title: Hindsight policy gradients, Abstract: A reinforcement learning agent that needs to pursue different goals across episodes requires a goal-conditional policy. In addition to their potential to generalize desirable behavior to unseen goals, such policies may also enable higher-level planning based on subgoals. In sparse-reward environments, the capacity to exploit information about the degree to which an arbitrary goal has been achieved while another goal was intended appears crucial to enable sample efficient learning. However, reinforcement learning agents have only recently been endowed with such capacity for hindsight. In this paper, we demonstrate how hindsight can be introduced to policy gradient methods, generalizing this idea to a broad class of successful algorithms. Our experiments on a diverse selection of sparse-reward environments show that hindsight leads to a remarkable increase in sample efficiency.

Title: Electron-Phonon Interaction in Ternary Rare-Earth Copper Antimonides LaCuSb2 and La(Cu0.8Ag0.2)Sb2 probed by Yanson Point-Contact Spectroscopy, Abstract: Investigation of the electron-phonon interaction (EPI) in LaCuSb2 and La(Cu0.8Ag0.2)Sb2 compounds by Yanson point-contact spectroscopy (PCS) has been carried out. Point-contact spectra display a pronounced broad maximum in the range of 10÷20 mV caused by EPI. Variation of the position of this maximum is likely connected with anisotropic phonon spectrum in these layered compounds. The absence of phonon features after the main maximum allows the assessment of the Debye energy of about 40 meV. The EPI constant for the LaCuSb2 compound was estimated to be {\lambda}=0.2+/-0.03. A zero-bias minimum in differential resistance for the latter compound is observed for some point contacts, which vanishes at about 6 K, pointing to the formation of superconducting phase under point contact, while superconducting critical temperature of the bulk sample is only 1K.

Title: Time-dependent linear-response variational Monte Carlo, Abstract: We present the extension of variational Monte Carlo (VMC) to the calculation of electronic excitation energies and oscillator strengths using time-dependent linear-response theory. By exploiting the analogy existing between the linear method for wave-function optimisation and the generalised eigenvalue equation of linear-response theory, we formulate the equations of linear-response VMC (LR-VMC). This LR-VMC approach involves the first-and second-order derivatives of the wave function with respect to the parameters. We perform first tests of the LR-VMC method within the Tamm-Dancoff approximation using single-determinant Jastrow-Slater wave functions with different Slater basis sets on some singlet and triplet excitations of the beryllium atom. Comparison with reference experimental data and with configuration-interaction-singles (CIS) results shows that LR-VMC generally outperforms CIS for excitation energies and is thus a promising approach for calculating electronic excited-state properties of atoms and molecules.

Title: Replica analysis of overfitting in regression models for time-to-event data, Abstract: Overfitting, which happens when the number of parameters in a model is too large compared to the number of data points available for determining these parameters, is a serious and growing problem in survival analysis. While modern medicine presents us with data of unprecedented dimensionality, these data cannot yet be used effectively for clinical outcome prediction. Standard error measures in maximum likelihood regression, such as p-values and z-scores, are blind to overfitting, and even for Cox's proportional hazards model (the main tool of medical statisticians), one finds in literature only rules of thumb on the number of samples required to avoid overfitting. In this paper we present a mathematical theory of overfitting in regression models for time-to-event data, which aims to increase our quantitative understanding of the problem and provide practical tools with which to correct regression outcomes for the impact of overfitting. It is based on the replica method, a statistical mechanical technique for the analysis of heterogeneous many-variable systems that has been used successfully for several decades in physics, biology, and computer science, but not yet in medical statistics. We develop the theory initially for arbitrary regression models for time-to-event data, and verify its predictions in detail for the popular Cox model.

Title: Spinor analysis, Abstract: "Let us call the novel quantities which, in addition to the vectors and tensors, have appeared in the quantum mechanics of the spinning electron, and which in the case of the Lorentz group are quite differently transformed from tensors, as spinors for short. Is there no spinor analysis that every physicist can learn, such as tensor analysis, and with the aid of which all the possible spinors can be formed, and secondly, all the invariant equations in which spinors occur?" So Mr Ehrenfest asked me and the answer will be given below.

Title: Identifiability of phylogenetic parameters from k-mer data under the coalescent, Abstract: Distances between sequences based on their $k$-mer frequency counts can be used to reconstruct phylogenies without first computing a sequence alignment. Past work has shown that effective use of k-mer methods depends on 1) model-based corrections to distances based on $k$-mers and 2) breaking long sequences into blocks to obtain repeated trials from the sequence-generating process. Good performance of such methods is based on having many high-quality blocks with many homologous sites, which can be problematic to guarantee a priori. Nature provides natural blocks of sequences into homologous regions---namely, the genes. However, directly using past work in this setting is problematic because of possible discordance between different gene trees and the underlying species tree. Using the multispecies coalescent model as a basis, we derive model-based moment formulas that involve the divergence times and the coalescent parameters. From this setting, we prove identifiability results for the tree and branch length parameters under the Jukes-Cantor model of sequence mutations.

Title: Using Transfer Learning for Image-Based Cassava Disease Detection, Abstract: Cassava is the third largest source of carbohydrates for human food in the world but is vulnerable to virus diseases, which threaten to destabilize food security in sub-Saharan Africa. Novel methods of cassava disease detection are needed to support improved control which will prevent this crisis. Image recognition offers both a cost effective and scalable technology for disease detection. New transfer learning methods offer an avenue for this technology to be easily deployed on mobile devices. Using a dataset of cassava disease images taken in the field in Tanzania, we applied transfer learning to train a deep convolutional neural network to identify three diseases and two types of pest damage (or lack thereof). The best trained model accuracies were 98% for brown leaf spot (BLS), 96% for red mite damage (RMD), 95% for green mite damage (GMD), 98% for cassava brown streak disease (CBSD), and 96% for cassava mosaic disease (CMD). The best model achieved an overall accuracy of 93% for data not used in the training process. Our results show that the transfer learning approach for image recognition of field images offers a fast, affordable, and easily deployable strategy for digital plant disease detection.

Title: Discovery of Extreme [OIII]+H$β$ Emitting Galaxies Tracing an Overdensity at z~3.5 in CDF-South, Abstract: Using deep multi-wavelength photometry of galaxies from ZFOURGE, we group galaxies at $2.5<z<4.0$ by the shape of their spectral energy distributions (SEDs). We identify a population of galaxies with excess emission in the $K_s$-band, which corresponds to [OIII]+H$\beta$ emission at $2.95<z<3.65$. This population includes 78% of the bluest galaxies with UV slopes steeper than $\beta = -2$. We de-redshift and scale this photometry to build two composite SEDs, enabling us to measure equivalent widths of these Extreme [OIII]+H$\beta$ Emission Line Galaxies (EELGs) at $z\sim3.5$. We identify 60 galaxies that comprise a composite SED with [OIII]+H$\beta$ rest-frame equivalent width of $803\pm228$\AA\ and another 218 galaxies in a composite SED with equivalent width of $230\pm90$\AA. These EELGs are analogous to the `green peas' found in the SDSS, and are thought to be undergoing their first burst of star formation due to their blue colors ($\beta < -1.6$), young ages ($\log(\rm{age}/yr)\sim7.2$), and low dust attenuation values. Their strong nebular emission lines and compact sizes (typically $\sim1.4$ kpc) are consistent with the properties of the star-forming galaxies possibly responsible for reionizing the universe at $z>6$. Many of the EELGs also exhibit Lyman-$\alpha$ emission. Additionally, we find that many of these sources are clustered in an overdensity in the Chandra Deep Field South, with five spectroscopically confirmed members at $z=3.474 \pm 0.004$. The spatial distribution and photometric redshifts of the ZFOURGE population further confirm the overdensity highlighted by the EELGs.

Title: Learning to attend in a brain-inspired deep neural network, Abstract: Recent machine learning models have shown that including attention as a component results in improved model accuracy and interpretability, despite the concept of attention in these approaches only loosely approximating the brain's attention mechanism. Here we extend this work by building a more brain-inspired deep network model of the primate ATTention Network (ATTNet) that learns to shift its attention so as to maximize the reward. Using deep reinforcement learning, ATTNet learned to shift its attention to the visual features of a target category in the context of a search task. ATTNet's dorsal layers also learned to prioritize these shifts of attention so as to maximize success of the ventral pathway classification and receive greater reward. Model behavior was tested against the fixations made by subjects searching images for the same cued category. Both subjects and ATTNet showed evidence for attention being preferentially directed to target goals, behaviorally measured as oculomotor guidance to targets. More fundamentally, ATTNet learned to shift its attention to target like objects and spatially route its visual inputs to accomplish the task. This work makes a step toward a better understanding of the role of attention in the brain and other computational systems.

Title: Anisotropic functional Laplace deconvolution, Abstract: In the present paper we consider the problem of estimating a three-dimensional function $f$ based on observations from its noisy Laplace convolution. Our study is motivated by the analysis of Dynamic Contrast Enhanced (DCE) imaging data. We construct an adaptive wavelet-Laguerre estimator of $f$, derive minimax lower bounds for the $L^2$-risk when $f$ belongs to a three-dimensional Laguerre-Sobolev ball and demonstrate that the wavelet-Laguerre estimator is adaptive and asymptotically near-optimal in a wide range of Laguerre-Sobolev spaces. We carry out a limited simulations study and show that the estimator performs well in a finite sample setting. Finally, we use the technique for the solution of the Laplace deconvolution problem on the basis of DCE Computerized Tomography data.

Title: Parameter Sharing Deep Deterministic Policy Gradient for Cooperative Multi-agent Reinforcement Learning, Abstract: Deep reinforcement learning for multi-agent cooperation and competition has been a hot topic recently. This paper focuses on cooperative multi-agent problem based on actor-critic methods under local observations settings. Multi agent deep deterministic policy gradient obtained state of art results for some multi-agent games, whereas, it cannot scale well with growing amount of agents. In order to boost scalability, we propose a parameter sharing deterministic policy gradient method with three variants based on neural networks, including actor-critic sharing, actor sharing and actor sharing with partially shared critic. Benchmarks from rllab show that the proposed method has advantages in learning speed and memory efficiency, well scales with growing amount of agents, and moreover, it can make full use of reward sharing and exchangeability if possible.

Title: On a Neumann-type series for modified Bessel functions of the first kind, Abstract: In this paper, we are interested in a Neumann-type series for modified Bessel functions of the first kind which arises in the study of Dunkl operators associated with dihedral groups and as an instance of the Laguerre semigroup constructed by Ben Said-Kobayashi-Orsted. We first revisit the particular case corresponding to the group of square-preserving symmetries for which we give two new and different proofs other than the existing ones. The first proof uses the expansion of powers in a Neumann series of Bessel functions while the second one is based on a quadratic transformation for the Gauss hypergeometric function and opens the way to derive further expressions when the orders of the underlying dihedral groups are powers of two. More generally, we give another proof of De Bie \& al formula expressing this series as a $\Phi_2$-Horn confluent hypergeometric function. In the course of proving, we shed the light on the occurrence of multiple angles in their formula through elementary symmetric functions, and get a new representation of Gegenbauer polynomials.

Title: A Modern Search for Wolf-Rayet Stars in the Magellanic Clouds. III. A Third Year of Discoveries, Abstract: For the past three years we have been conducting a survey for WR stars in the Large and Small Magellanic Clouds (LMC, SMC). Our previous work has resulted in the discovery of a new type of WR star in the LMC, which we are calling WN3/O3. These stars have the emission-line properties of a WN3 star (strong N V but no N IV), plus the absorption-line properties of an O3 star (Balmer hydrogen plus Pickering He II but no He I). Yet these stars are 15x fainter than an O3 V star would be by itself, ruling out these being WN3+O3 binaries. Here we report the discovery of two more members of this class, bringing the total number of these objects to 10, 6.5% of the LMC's total WR population. The optical spectra of nine of these WN3/O3s are virtually indistinguishable from each other, but one of the newly found stars is significantly different, showing a lower excitation emission and absorption spectrum (WN4/O4-ish). In addition, we have newly classified three unusual Of-type stars, including one with a strong C III 4650 line, and two rapidly rotating "Oef" stars. We also "rediscovered" a low mass x-ray binary, RX J0513.9-6951, and demonstrate its spectral variability. Finally, we discuss the spectra of ten low priority WR candidates that turned out not to have He II emission. These include both a Be star and a B[e] star.

Title: Mathematical modeling of Zika disease in pregnant women and newborns with microcephaly in Brazil, Abstract: We propose a new mathematical model for the spread of Zika virus. Special attention is paid to the transmission of microcephaly. Numerical simulations show the accuracy of the model with respect to the Zika outbreak occurred in Brazil.