Datasets:

bhujith10

/

multi_class_classification_dataset

like 0

Title: Universal Constraints on the Location of Extrema of Eigenfunctions of Non-Local Schrödinger Operators, Abstract: We derive a lower bound on the location of global extrema of eigenfunctions for a large class of non-local Schrödinger operators in convex domains under Dirichlet exterior conditions, featuring the symbol of the kinetic term, the strength of the potential, and the corresponding eigenvalue, and involving a new universal constant. We show a number of probabilistic and spectral geometric implications, and derive a Faber-Krahn type inequality for non-local operators. Our study also extends to potentials with compact support, and we establish bounds on the location of extrema relative to the boundary edge of the support or level sets around minima of the potential.

Title: InScript: Narrative texts annotated with script information, Abstract: This paper presents the InScript corpus (Narrative Texts Instantiating Script structure). InScript is a corpus of 1,000 stories centered around 10 different scenarios. Verbs and noun phrases are annotated with event and participant types, respectively. Additionally, the text is annotated with coreference information. The corpus shows rich lexical variation and will serve as a unique resource for the study of the role of script knowledge in natural language processing.

Title: Semimetallic and charge-ordered $α$-(BEDT-TTF)$_2$I$_3$: on the role of disorder in dc transport and dielectric properties, Abstract: $\alpha$-(BEDT-TTF)$_2$I$_3$ is a prominent example of charge ordering among organic conductors. In this work we explore the details of transport within the charge-ordered as well as semimetallic phase at ambient pressure. In the high-temperature semimetallic phase, the mobilities and concentrations of both electrons and holes conspire in such a way to create an almost temperature-independent conductivity as well as a low Hall effect. We explain these phenomena as a consequence of a predominantly inter-pocket scattering which equalizes mobilities of the two types of charge carriers. At low temperatures, within the insulating charge-ordered phase two channels of conduction can be discerned: a temperature-dependent activation which follows the mean-field behavior, and a nearest-neighbor hopping contribution. Together with negative magnetoresistance, the latter relies on the presence of disorder. The charge-ordered phase also features a prominent dielectric peak which bears a similarity to relaxor ferroelectrics. Its dispersion is determined by free-electron screening and pushed by disorder well below the transition temperature. The source of this disorder can be found in the anion layers which randomly perturb BEDT-TTF molecules through hydrogen bonds.

Title: Robust Photometric Stereo Using Learned Image and Gradient Dictionaries, Abstract: Photometric stereo is a method for estimating the normal vectors of an object from images of the object under varying lighting conditions. Motivated by several recent works that extend photometric stereo to more general objects and lighting conditions, we study a new robust approach to photometric stereo that utilizes dictionary learning. Specifically, we propose and analyze two approaches to adaptive dictionary regularization for the photometric stereo problem. First, we propose an image preprocessing step that utilizes an adaptive dictionary learning model to remove noise and other non-idealities from the image dataset before estimating the normal vectors. We also propose an alternative model where we directly apply the adaptive dictionary regularization to the normal vectors themselves during estimation. We study the practical performance of both methods through extensive simulations, which demonstrate the state-of-the-art performance of both methods in the presence of noise.

Title: Toward Low-Flying Autonomous MAV Trail Navigation using Deep Neural Networks for Environmental Awareness, Abstract: We present a micro aerial vehicle (MAV) system, built with inexpensive off-the-shelf hardware, for autonomously following trails in unstructured, outdoor environments such as forests. The system introduces a deep neural network (DNN) called TrailNet for estimating the view orientation and lateral offset of the MAV with respect to the trail center. The DNN-based controller achieves stable flight without oscillations by avoiding overconfident behavior through a loss function that includes both label smoothing and entropy reward. In addition to the TrailNet DNN, the system also utilizes vision modules for environmental awareness, including another DNN for object detection and a visual odometry component for estimating depth for the purpose of low-level obstacle detection. All vision systems run in real time on board the MAV via a Jetson TX1. We provide details on the hardware and software used, as well as implementation details. We present experiments showing the ability of our system to navigate forest trails more robustly than previous techniques, including autonomous flights of 1 km.

Title: Origin of Charge Separation at Organic Photovoltaic Heterojunctions: A Mesoscale Quantum Mechanical View, Abstract: The high efficiency of charge generation within organic photovoltaic blends apparently contrasts with the strong "classical" attraction between newly formed electron-hole pairs. Several factors have been identified as possible facilitators of charge dissociation, such as quantum mechanical coherence and delocalization, structural and energetic disorder, built-in electric fields, nanoscale intermixing of the donor and acceptor components of the blends. Our mesoscale quantum-chemical model allows an unbiased assessment of their relative importance, through excited-state calculations on systems containing thousands of donor and acceptor sites. The results on several model heterojunctions confirm that the classical model severely overestimates the binding energy of the electron-hole pairs, produced by vertical excitation from the electronic ground state. Using physically sensible parameters for the individual materials, we find that the quantum mechanical energy difference between the lowest interfacial charge transfer states and the fully separated electron and hole is of the order of the thermal energy.

Title: Spectral Decimation for Families of Self-Similar Symmetric Laplacians on the Sierpinski Gasket, Abstract: We construct a one-parameter family of Laplacians on the Sierpinski Gasket that are symmetric and self-similar for the 9-map iterated function system obtained by iterating the standard 3-map iterated function system. Our main result is the fact that all these Laplacians satisfy a version of spectral decimation that builds a precise catalog of eigenvalues and eigenfunctions for any choice of the parameter. We give a number of applications of this spectral decimation. We also prove analogous results for fractal Laplacians on the unit Interval, and this yields an analogue of the classical Sturm-Liouville theory for the eigenfunctions of these one-dimensional Laplacians.

Title: Identification of multi-object dynamical systems: consistency and Fisher information, Abstract: Learning the model parameters of a multi-object dynamical system from partial and perturbed observations is a challenging task. Despite recent numerical advancements in learning these parameters, theoretical guarantees are extremely scarce. In this article, we study the identifiability of these parameters and the consistency of the corresponding maximum likelihood estimate (MLE) under assumptions on the different components of the underlying multi-object system. In order to understand the impact of the various sources of observation noise on the ability to learn the model parameters, we study the asymptotic variance of the MLE through the associated Fisher information matrix. For example, we show that specific aspects of the multi-target tracking (MTT) problem such as detection failures and unknown data association lead to a loss of information which is quantified in special cases of interest.

Title: Elliptic curves maximal over extensions of finite base fields, Abstract: Given an elliptic curve $E$ over a finite field $\mathbb{F}_q$ we study the finite extensions $\mathbb{F}_{q^n}$ of $\mathbb{F}_q$ such that the number of $\mathbb{F}_{q^n}$-rational points on $E$ attains the Hasse upper bound. We obtain an upper bound on the degree $n$ for $E$ ordinary using an estimate for linear forms in logarithms, which allows us to compute the pairs of isogeny classes of such curves and degree $n$ for small $q$. Using a consequence of Schmidt's Subspace Theorem, we improve the upper bound to $n\leq 11$ for sufficiently large $q$. We also show that there are infinitely many isogeny classes of ordinary elliptic curves with $n=3$.

Title: CURE: Curvature Regularization For Missing Data Recovery, Abstract: Missing data recovery is an important and yet challenging problem in imaging and data science. Successful models often adopt certain carefully chosen regularization. Recently, the low dimension manifold model (LDMM) was introduced by S.Osher et al. and shown effective in image inpainting. They observed that enforcing low dimensionality on image patch manifold serves as a good image regularizer. In this paper, we observe that having only the low dimension manifold regularization is not enough sometimes, and we need smoothness as well. For that, we introduce a new regularization by combining the low dimension manifold regularization with a higher order Curvature Regularization, and we call this new regularization CURE for short. The key step of solving CURE is to solve a biharmonic equation on a manifold. We further introduce a weighted version of CURE, called WeCURE, in a similar manner as the weighted nonlocal Laplacian (WNLL) method. Numerical experiments for image inpainting and semi-supervised learning show that the proposed CURE and WeCURE significantly outperform LDMM and WNLL respectively.

Title: Nonlinear parametric excitation effect induces stability transitions in swimming direction of flexible superparamagnetic microswimmers, Abstract: Microscopic artificial swimmers have recently become highly attractive due to their promising potential for biomedical applications. The pioneering work of Dreyfus et al (2005) has demonstrated the motion of a microswimmer with an undulating chain of superparamagnetic beads, which is actuated by an oscillating external magnetic field. Interestingly, it has also been theoretically predicted that the swimming direction of this swimmer will undergo a $90^\circ$-transition when the magnetic field's oscillations amplitude is increased above a critical value of $\sqrt{2}$. In this work, we further investigate this transition both theoretically and experimentally by using numerical simulations and presenting a novel flexible microswimmer with a superparamagnetic head. We realize the $90^\circ$-transition in swimming direction, prove that this effect depends on both frequency and amplitude of the oscillating magnetic field, and demonstrate the existence of an optimal amplitude, under which, maximal swimming speed can be achieved. By asymptotically analyzing the dynamic motion of microswimmer with a minimal two-link model, we reveal that the stability transitions representing the changes in the swimming direction are induced by the effect of nonlinear parametric excitation.

Title: $L^p$ Mapping Properties for the Cauchy-Riemann Equations on Lipschitz Domains Admitting Subelliptic Estimates, Abstract: We show that on bounded Lipschitz pseudoconvex domains that admit good weight functions the $\overline{\partial}$-Neumann operators $N_q, \overline{\partial}^* N_{q}$, and $\overline{\partial} N_{q}$ are bounded on $L^p$ spaces for some values of $p$ greater than 2.

Title: Fast dose optimization for rotating shield brachytherapy, Abstract: Purpose: To provide a fast computational method, based on the proximal graph solver (POGS) - a convex optimization solver using the alternating direction method of multipliers (ADMM), for calculating an optimal treatment plan in rotating shield brachytherapy (RSBT). RSBT treatment planning has more degrees of freedom than conventional high-dose-rate brachytherapy (HDR-BT) due to the addition of emission direction, and this necessitates a fast optimization technique to enable clinical usage. // Methods: The multi-helix RSBT (H-RSBT) delivery technique was considered with five representative cervical cancer patients. Treatment plans were generated for all patients using the POGS method and the previously considered commercial solver IBM CPLEX. The rectum, bladder, sigmoid, high-risk clinical target volume (HR-CTV), and HR-CTV boundary were the structures considered in our optimization problem, called the asymmetric dose-volume optimization with smoothness control. Dose calculation resolution was 1x1x3 mm^3 for all cases. The H-RSBT applicator has 6 helices, with 33.3 mm of translation along the applicator per helical rotation and 1.7 mm spacing between dwell positions, yielding 17.5 degree emission angle spacing per 5 mm along the applicator.// Results: For each patient, HR-CTV D90, HR-CTV D100, rectum D2cc, sigmoid D2cc, and bladder D2cc matched within 1% for CPLEX and POGS. Also, we obtained similar EQD2 figures between CPLEX and POGS. POGS was around 18 times faster than CPLEX. Over all patients, total optimization times were 32.1-65.4 seconds for CPLEX and 2.1-3.9 seconds for POGS. // Conclusions: POGS substantially reduced treatment plan optimization time around 18 times for RSBT with similar HR-CTV D90, OAR D2cc values, and EQD2 figure relative to CPLEX, which is significant progress toward clinical translation of RSBT. POGS is also applicable to conventional HDR-BT.

Title: 3D Modeling of Electric Fields in the LUX Detector, Abstract: This work details the development of a three-dimensional (3D) electric field model for the LUX detector. The detector took data during two periods of searching for weakly interacting massive particle (WIMP) searches. After the first period completed, a time-varying non-uniform negative charge developed in the polytetrafluoroethylene (PTFE) panels that define the radial boundary of the detector's active volume. This caused electric field variations in the detector in time, depth and azimuth, generating an electrostatic radially-inward force on electrons on their way upward to the liquid surface. To map this behavior, 3D electric field maps of the detector's active volume were built on a monthly basis. This was done by fitting a model built in COMSOL Multiphysics to the uniformly distributed calibration data that were collected on a regular basis. The modeled average PTFE charge density increased over the course of the exposure from -3.6 to $-5.5~\mu$C/m$^2$. From our studies, we deduce that the electric field magnitude varied while the mean value of the field of $\sim200$~V/cm remained constant throughout the exposure. As a result of this work the varying electric fields and their impact on event reconstruction and discrimination were successfully modeled.

Title: Optimal Communication Strategies in Networked Cyber-Physical Systems with Adversarial Elements, Abstract: This paper studies optimal communication and coordination strategies in cyber-physical systems for both defender and attacker within a game-theoretic framework. We model the communication network of a cyber-physical system as a sensor network which involves one single Gaussian source observed by many sensors, subject to additive independent Gaussian observation noises. The sensors communicate with the estimator over a coherent Gaussian multiple access channel. The aim of the receiver is to reconstruct the underlying source with minimum mean squared error. The scenario of interest here is one where some of the sensors are captured by the attacker and they act as the adversary (jammer): they strive to maximize distortion. The receiver (estimator) knows the captured sensors but still cannot simply ignore them due to the multiple access channel, i.e., the outputs of all sensors are summed to generate the estimator input. We show that the ability of transmitter sensors to secretly agree on a random event, that is "coordination", plays a key role in the analysis...

Title: First- and Second-Order Models of Recursive Arithmetics, Abstract: We study a quadruple of interrelated subexponential subsystems of arithmetic WKL$_0^-$, RCA$^-_0$, I$\Delta_0$, and $\Delta$RA$_1$, which complement the similarly related quadruple WKL$_0$, RCA$_0$, I$\Sigma_1$, and PRA studied by Simpson, and the quadruple WKL$_0^\ast$, RCA$_0^\ast$, I$\Delta_0$(exp), and EFA studied by Simpson and Smith. We then explore the space of subexponential arithmetic theories between I$\Delta_0$ and I$\Delta_0$(exp). We introduce and study first- and second-order theories of recursive arithmetic $A$RA$_1$ and $A$RA$_2$ capable of characterizing various computational complexity classes and based on function algebras $A$, studied by Clote and others.

Title: Land Cover Classification via Multi-temporal Spatial Data by Recurrent Neural Networks, Abstract: Nowadays, modern earth observation programs produce huge volumes of satellite images time series (SITS) that can be useful to monitor geographical areas through time. How to efficiently analyze such kind of information is still an open question in the remote sensing field. Recently, deep learning methods proved suitable to deal with remote sensing data mainly for scene classification (i.e. Convolutional Neural Networks - CNNs - on single images) while only very few studies exist involving temporal deep learning approaches (i.e Recurrent Neural Networks - RNNs) to deal with remote sensing time series. In this letter we evaluate the ability of Recurrent Neural Networks, in particular the Long-Short Term Memory (LSTM) model, to perform land cover classification considering multi-temporal spatial data derived from a time series of satellite images. We carried out experiments on two different datasets considering both pixel-based and object-based classification. The obtained results show that Recurrent Neural Networks are competitive compared to state-of-the-art classifiers, and may outperform classical approaches in presence of low represented and/or highly mixed classes. We also show that using the alternative feature representation generated by LSTM can improve the performances of standard classifiers.

Title: Disentangling by Factorising, Abstract: We define and address the problem of unsupervised learning of disentangled representations on data generated from independent factors of variation. We propose FactorVAE, a method that disentangles by encouraging the distribution of representations to be factorial and hence independent across the dimensions. We show that it improves upon $\beta$-VAE by providing a better trade-off between disentanglement and reconstruction quality. Moreover, we highlight the problems of a commonly used disentanglement metric and introduce a new metric that does not suffer from them.

Title: Symmetries and synchronization in multilayer random networks, Abstract: In the light of the recently proposed scenario of asymmetry-induced synchronization (AISync), in which dynamical uniformity and consensus in a distributed system would demand certain asymmetries in the underlying network, we investigate here the influence of some regularities in the interlayer connection patterns on the synchronization properties of multilayer random networks. More specifically, by considering a Stuart-Landau model of complex oscillators with random frequencies, we report for multilayer networks a dynamical behavior that could be also classified as a manifestation of AISync. We show, namely, that the presence of certain symmetries in the interlayer connection pattern tends to diminish the synchronization capability of the whole network or, in other words, asymmetries in the interlayer connections would enhance synchronization in such structured networks. Our results might help the understanding not only of the AISync mechanism itself, but also its possible role in the determination of the interlayer connection pattern of multilayer and other structured networks with optimal synchronization properties.

Title: Application of Decision Rules for Handling Class Imbalance in Semantic Segmentation, Abstract: As part of autonomous car driving systems, semantic segmentation is an essential component to obtain a full understanding of the car's environment. One difficulty, that occurs while training neural networks for this purpose, is class imbalance of training data. Consequently, a neural network trained on unbalanced data in combination with maximum a-posteriori classification may easily ignore classes that are rare in terms of their frequency in the dataset. However, these classes are often of highest interest. We approach such potential misclassifications by weighting the posterior class probabilities with the prior class probabilities which in our case are the inverse frequencies of the corresponding classes in the training dataset. More precisely, we adopt a localized method by computing the priors pixel-wise such that the impact can be analyzed at pixel level as well. In our experiments, we train one network from scratch using a proprietary dataset containing 20,000 annotated frames of video sequences recorded from street scenes. The evaluation on our test set shows an increase of average recall with regard to instances of pedestrians and info signs by $25\%$ and $23.4\%$, respectively. In addition, we significantly reduce the non-detection rate for instances of the same classes by $61\%$ and $38\%$.

Title: Muchnik degrees and cardinal characteristics, Abstract: We provide a pair of dual results, each stating the coincidence of highness properties from computability theory. We provide an analogous pair of dual results on the coincidence of cardinal characteristics within ZFC. A mass problem is a set of functions on $\omega$. For mass problems $\mathcal C, \mathcal D$, one says that $\mathcal C$ is Muchnik reducible to $\mathcal D$ if each function in $\mathcal D$ computes a function in $\mathcal C$. In this paper we view highness properties as mass problems, and compare them with respect to Muchnik reducibility and its uniform strengthening, Medvedev reducibility. Let $\mathcal D(p)$ be the mass problem of infinite bit sequences $y$ (i.e., 0,1 valued functions) such that for each computable bit sequence $x$, the asymptotic lower density $\underline \rho$ of the agreement bit sequence $x \leftrightarrow y$ is at most $p$ (this sequence takes the value 1 at a bit position iff $x$ and $y$ agree). We show that all members of this family of mass problems parameterized by a real $p$ with $0 < p<1/2 $ have the same complexity in the sense of Muchnik reducibility. This also yields a new version of Monin's affirmative answer to the "Gamma question", whether $\Gamma(A)< 1/2$ implies $\Gamma(A)=0$ for each Turing oracle $A$. We also show, together with Joseph Miller, that for any order function~$g$ there exists a faster growing order function $h $ such that $\mathrm{IOE}(g) $ is strictly Muchnik below $\mathrm{IOE}(h)$. We study cardinal characteristics analogous to the highness properties above. For instance, $\mathfrak d (p)$ is the least size of a set $G$ of bit sequences so that for each bit sequence $x$ there is a bit sequence $y$ in $G$ so that $\underline \rho (x \leftrightarrow y) >p$. We prove within ZFC all the coincidences of cardinal characteristics that are the analogs of the results above.

Title: Existence of solutions for a semirelativistic Hartree equation with unbounded potentials, Abstract: We prove the existence of a solution to the semirelativistic Hartree equation $$\sqrt{-\Delta+m^2}u+ V(x) u = A(x)\left( W * |u|^p \right) |u|^{p-2}u $$ under suitable growth assumption on the potential functions $V$ and $A$. In particular, both can be unbounded from above.

Title: Lattice Operations on Terms over Similar Signatures, Abstract: Unification and generalization are operations on two terms computing respectively their greatest lower bound and least upper bound when the terms are quasi-ordered by subsumption up to variable renaming (i.e., $t_1\preceq t_2$ iff $t_1 = t_2\sigma$ for some variable substitution $\sigma$). When term signatures are such that distinct functor symbols may be related with a fuzzy equivalence (called a similarity), these operations can be formally extended to tolerate mismatches on functor names and/or arity or argument order. We reformulate and extend previous work with a declarative approach defining unification and generalization as sets of axioms and rules forming a complete constraint-normalization proof system. These include the Reynolds-Plotkin term-generalization procedures, Maria Sessa's "weak" unification with partially fuzzy signatures and its corresponding generalization, as well as novel extensions of such operations to fully fuzzy signatures (i.e., similar functors with possibly different arities). One advantage of this approach is that it requires no modification of the conventional data structures for terms and substitutions. This and the fact that these declarative specifications are efficiently executable conditional Horn-clauses offers great practical potential for fuzzy information-handling applications.

Title: Towards Physically Safe Reinforcement Learning under Supervision, Abstract: This paper addresses the question of how a previously available control policy $\pi_s$ can be used as a supervisor to more quickly and safely train a new learned control policy $\pi_L$ for a robot. A weighted average of the supervisor and learned policies is used during trials, with a heavier weight initially on the supervisor, in order to allow safe and useful physical trials while the learned policy is still ineffective. During the process, the weight is adjusted to favor the learned policy. As weights are adjusted, the learned network must compensate so as to give safe and reasonable outputs under the different weights. A pioneer network is introduced that pre-learns a policy that performs similarly to the current learned policy under the planned next step for new weights; this pioneer network then replaces the currently learned network in the next set of trials. Experiments in OpenAI Gym demonstrate the effectiveness of the proposed method.

Title: Complex Valued Risk Diversification, Abstract: Risk diversification is one of the dominant concerns for portfolio managers. Various portfolio constructions have been proposed to minimize the risk of the portfolio under some constrains including expected returns. We propose a portfolio construction method that incorporates the complex valued principal component analysis into the risk diversification portfolio construction. The proposed method is verified to outperform the conventional risk parity and risk diversification portfolio constructions.

Title: An Overview on Application of Machine Learning Techniques in Optical Networks, Abstract: Today's telecommunication networks have become sources of enormous amounts of widely heterogeneous data. This information can be retrieved from network traffic traces, network alarms, signal quality indicators, users' behavioral data, etc. Advanced mathematical tools are required to extract meaningful information from these data and take decisions pertaining to the proper functioning of the networks from the network-generated data. Among these mathematical tools, Machine Learning (ML) is regarded as one of the most promising methodological approaches to perform network-data analysis and enable automated network self-configuration and fault management. The adoption of ML techniques in the field of optical communication networks is motivated by the unprecedented growth of network complexity faced by optical networks in the last few years. Such complexity increase is due to the introduction of a huge number of adjustable and interdependent system parameters (e.g., routing configurations, modulation format, symbol rate, coding schemes, etc.) that are enabled by the usage of coherent transmission/reception technologies, advanced digital signal processing and compensation of nonlinear effects in optical fiber propagation. In this paper we provide an overview of the application of ML to optical communications and networking. We classify and survey relevant literature dealing with the topic, and we also provide an introductory tutorial on ML for researchers and practitioners interested in this field. Although a good number of research papers have recently appeared, the application of ML to optical networks is still in its infancy: to stimulate further work in this area, we conclude the paper proposing new possible research directions.

Title: Study of Minor Actinides Transmutation in PWR MOX fuel, Abstract: The management of long-lived radionuclides in spent fuel is a key issue to achieve the closed nuclear fuel cycle and the sustainable development of nuclear energy. Partitioning-Transmutation is supposed to be an efficient method to treat the long-lived radionuclides in spent fuel. Some Minor Actinides (MAs) have very long half-lives among the radionuclides in the spent fuel. Accordingly, the study of MAs transmutation is a significant work for the post-processing of spent fuel. In the present work, the transmutations in Pressurized Water Reactor (PWR) mixed oxide (MOX) fuel are investigated through the Monte Carlo based code RMC. Two kinds of MAs, $^{237}$Np and five MAs ($^{237}$Np, $^{241}$Am, $^{243}$Am, $^{244}$Cm and $^{245}$Cm) are incorporated homogeneously into the MOX fuel assembly. The transmutation of MAs is simulated with different initial MOX concentrations. The results indicate an overall nice efficiency of transmutation in both initial MOX concentrations, especially for the two kinds of MAs primarily generated in the UOX fuel, $^{237}$Np and $^{241}$Am. In addition, the inclusion of $^{237}$Np in MOX has no large influence for other MAs, while the transmutation efficiency of $^{237}$Np is excellent. The transmutation of MAs in MOX fuel depletion is expected to be a new, efficient nuclear spent fuel management method for the future nuclear power generation.

Title: Undesired parking spaces and contractible pieces of the noncrossing partition link, Abstract: There are two natural simplicial complexes associated to the noncrossing partition lattice: the order complex of the full lattice and the order complex of the lattice with its bounding elements removed. The latter is a complex that we call the noncrossing partition link because it is the link of an edge in the former. The first author and his coauthors conjectured that various collections of simplices of the noncrossing partition link (determined by the undesired parking spaces in the corresponding parking functions) form contractible subcomplexes. In this article we prove their conjecture by combining the fact that the star of a simplex in a flag complex is contractible with the second author's theory of noncrossing hypertrees.

Title: Breaking the Nonsmooth Barrier: A Scalable Parallel Method for Composite Optimization, Abstract: Due to their simplicity and excellent performance, parallel asynchronous variants of stochastic gradient descent have become popular methods to solve a wide range of large-scale optimization problems on multi-core architectures. Yet, despite their practical success, support for nonsmooth objectives is still lacking, making them unsuitable for many problems of interest in machine learning, such as the Lasso, group Lasso or empirical risk minimization with convex constraints. In this work, we propose and analyze ProxASAGA, a fully asynchronous sparse method inspired by SAGA, a variance reduced incremental gradient algorithm. The proposed method is easy to implement and significantly outperforms the state of the art on several nonsmooth, large-scale problems. We prove that our method achieves a theoretical linear speedup with respect to the sequential version under assumptions on the sparsity of gradients and block-separability of the proximal term. Empirical benchmarks on a multi-core architecture illustrate practical speedups of up to 12x on a 20-core machine.

Title: Sparsity-promoting and edge-preserving maximum a posteriori estimators in non-parametric Bayesian inverse problems, Abstract: We consider the inverse problem of recovering an unknown functional parameter $u$ in a separable Banach space, from a noisy observation $y$ of its image through a known possibly non-linear ill-posed map ${\mathcal G}$. The data $y$ is finite-dimensional and the noise is Gaussian. We adopt a Bayesian approach to the problem and consider Besov space priors (see Lassas et al. 2009), which are well-known for their edge-preserving and sparsity-promoting properties and have recently attracted wide attention especially in the medical imaging community. Our key result is to show that in this non-parametric setup the maximum a posteriori (MAP) estimates are characterized by the minimizers of a generalized Onsager--Machlup functional of the posterior. This is done independently for the so-called weak and strong MAP estimates, which as we show coincide in our context. In addition, we prove a form of weak consistency for the MAP estimators in the infinitely informative data limit. Our results are remarkable for two reasons: first, the prior distribution is non-Gaussian and does not meet the smoothness conditions required in previous research on non-parametric MAP estimates. Second, the result analytically justifies existing uses of the MAP estimate in finite but high dimensional discretizations of Bayesian inverse problems with the considered Besov priors.

Title: Non-LTE line formation of Fe in late-type stars IV: Modelling of the solar centre-to-limb variation in 3D, Abstract: Our ability to model the shapes and strengths of iron lines in the solar spectrum is a critical test of the accuracy of the solar iron abundance, which sets the absolute zero-point of all stellar metallicities. We use an extensive 463-level Fe atom with new photoionisation cross-sections for FeI as well as quantum mechanical calculations of collisional excitation and charge transfer with neutral hydrogen; the latter effectively remove a free parameter that has hampered all previous line formation studies of Fe in non-local thermodynamic equilibrium (NLTE). For the first time, we use realistic 3D NLTE calculations of Fe for a quantitative comparison to solar observations. We confront our theoretical line profiles with observations taken at different viewing angles across the solar disk with the Swedish 1-m Solar Telescope. We find that 3D modelling well reproduces the observed centre-to-limb behaviour of spectral lines overall, but highlight aspects that may require further work, especially cross-sections for inelastic collisions with electrons. Our inferred solar iron abundance is log(eps(Fe))=7.48+-0.04.

Title: Transmission clusters in the HIV-1 epidemic among men who have sex with men in Montreal, Quebec, Canada, Abstract: Background. Several studies have used phylogenetics to investigate Human Immunodeficiency Virus (HIV) transmission among Men who have Sex with Men (MSMs) in Montreal, Quebec, Canada, revealing many transmission clusters. The Quebec HIV genotyping program sequence database now includes viral sequences from close to 4,000 HIV-positive individuals classified as MSMs. In this paper, we investigate clustering in those data by comparing results from several methods: the conventional Bayesian and maximum likelihood-bootstrap methods, and two more recent algorithms, DM-PhyClus, a Bayesian algorithm that produces a measure of uncertainty for proposed partitions, and the Gap Procedure, a fast distance-based approach. We estimate cluster growth by focusing on recent cases in the Primary HIV Infection (PHI) stage. Results. The analyses reveal considerable overlap between cluster estimates obtained from conventional methods. The Gap Procedure and DM-PhyClus rely on different cluster definitions and as a result, suggest moderately different partitions. All estimates lead to similar conclusions about cluster expansion: several large clusters have experienced sizeable growth, and a few new transmission clusters are likely emerging. Conclusions. The lack of a gold standard measure for clustering quality makes picking a best estimate among those proposed difficult. Work aiming to refine clustering criteria would be required to improve estimates. Nevertheless, the results unanimously stress the role that clusters play in promoting HIV incidence among MSMs.

Title: Conformal blocks attached to twisted groups, Abstract: The aim of this paper is to generalize the notion of conformal blocks to the situation in which the Lie algebra they are attached to is not defined over a field, but depends on covering data of curves. The result will be a sheaf of conformal blocks on the Hurwitz stack parametrizing Galois coverings of curves. Many features of the classical sheaves of conformal blocks are proved to hold in this more general setting, in particular the fusion rules, the propagation of vacua and the WZW connection.

Title: Statistical study on propagation characteristics of Omega signals (VLF) in magnetosphere detected by the Akebono satellite, Abstract: This paper shows a statistical analysis of 10.2 kHz Omega broadcasts of an artificial signal broadcast from ground stations, propagated in the plasmasphere, and detected using an automatic detection method we developed. We study the propagation patterns of the Omega signals to understand the propagation characteristics that are strongly affected by plasmaspheric electron density and the ambient magnetic field. We show the unique propagation patterns of the Omega 10.2 kHz signal when it was broadcast from two high-middle-latitude stations. We use about eight years of data captured by the Poynting flux analyzer subsystem on board the Akebono satellite from October 1989 to September 1997. We demonstrate that the signals broadcast from almost the same latitude (in geomagnetic coordinates) propagated differently depending on the geographic latitude. We also study propagation characteristics as a function of local time, season, and solar activity. The Omega signal tended to propagate farther on the nightside than on the dayside and was more widely distributed during winter than during summer. When solar activity was at maximum, the Omega signal propagated at a lower intensity level. In contrast, when solar activity was at minimum, the Omega signal propagated at a higher intensity and farther from the transmitter station.

Title: Thermally induced stresses in boulders on airless body surfaces, and implications for rock breakdown, Abstract: This work investigates the macroscopic thermomechanical behavior of lunar boulders by modeling their response to diurnal thermal forcing. Our results reveal a bimodal, spatiotemporally-complex stress response. During sunrise, stresses occur in the boulders' interiors that are associated with large-scale temperature gradients developed due to overnight cooling. During sunset, stresses occur at the boulders' exteriors due to the cooling and contraction of the surface. Both kinds of stresses are on the order of 10 MPa in 1 m boulders and decrease for smaller diameters, suggesting that larger boulders break down more quickly. Boulders <30 cm exhibit a weak response to thermal forcing, suggesting a threshold below which crack propagation may not occur. Boulders of any size buried by regolith are shielded from thermal breakdown. As boulders increase in size (>1 m), stresses increase to several 10s of MPa as the behavior of their surfaces approaches that of an infinite halfspace. As the thermal wave loses contact with the boulder interior, stresses become limited to the near-surface. This suggests that the survival time of a boulder is not only controlled by the amplitude of induced stress, but also by its diameter as compared to the diurnal skin depth. While stresses on the order of 10 MPa are enough to drive crack propagation in terrestrial environments, crack propagation rates in vacuum are not well constrained. We explore the relationship between boulder size, stress, and the direction of crack propagation, and discuss the implications for the relative breakdown rates and estimated lifetimes of boulders on airless body surfaces.

Title: Some observations about generalized quantifiers in logics of imperfect information, Abstract: We analyze the definitions of generalized quantifiers of imperfect information that have been proposed by F.Engström. We argue that these definitions are just embeddings of the first-order generalized quantifiers into team semantics, and fail to capture an adequate notion of team-theoretical generalized quantifier, save for the special cases in which the quantifiers are applied to flat formulas. We also criticize the meaningfulness of the monotone/nonmonotone distinction in this context. We make some proposals for a more adequate definition of generalized quantifiers of imperfect information.

Title: Predicting Demographics of High-Resolution Geographies with Geotagged Tweets, Abstract: In this paper, we consider the problem of predicting demographics of geographic units given geotagged Tweets that are composed within these units. Traditional survey methods that offer demographics estimates are usually limited in terms of geographic resolution, geographic boundaries, and time intervals. Thus, it would be highly useful to develop computational methods that can complement traditional survey methods by offering demographics estimates at finer geographic resolutions, with flexible geographic boundaries (i.e. not confined to administrative boundaries), and at different time intervals. While prior work has focused on predicting demographics and health statistics at relatively coarse geographic resolutions such as the county-level or state-level, we introduce an approach to predict demographics at finer geographic resolutions such as the blockgroup-level. For the task of predicting gender and race/ethnicity counts at the blockgroup-level, an approach adapted from prior work to our problem achieves an average correlation of 0.389 (gender) and 0.569 (race) on a held-out test dataset. Our approach outperforms this prior approach with an average correlation of 0.671 (gender) and 0.692 (race).

Title: Text Compression for Sentiment Analysis via Evolutionary Algorithms, Abstract: Can textual data be compressed intelligently without losing accuracy in evaluating sentiment? In this study, we propose a novel evolutionary compression algorithm, PARSEC (PARts-of-Speech for sEntiment Compression), which makes use of Parts-of-Speech tags to compress text in a way that sacrifices minimal classification accuracy when used in conjunction with sentiment analysis algorithms. An analysis of PARSEC with eight commercial and non-commercial sentiment analysis algorithms on twelve English sentiment data sets reveals that accurate compression is possible with (0%, 1.3%, 3.3%) loss in sentiment classification accuracy for (20%, 50%, 75%) data compression with PARSEC using LingPipe, the most accurate of the sentiment algorithms. Other sentiment analysis algorithms are more severely affected by compression. We conclude that significant compression of text data is possible for sentiment analysis depending on the accuracy demands of the specific application and the specific sentiment analysis algorithm used.

Title: Training large margin host-pathogen protein-protein interaction predictors, Abstract: Detection of protein-protein interactions (PPIs) plays a vital role in molecular biology. Particularly, infections are caused by the interactions of host and pathogen proteins. It is important to identify host-pathogen interactions (HPIs) to discover new drugs to counter infectious diseases. Conventional wet lab PPI prediction techniques have limitations in terms of large scale application and budget. Hence, computational approaches are developed to predict PPIs. This study aims to develop large margin machine learning models to predict interspecies PPIs with a special interest in host-pathogen protein interactions (HPIs). Especially, we focus on seeking answers to three queries that arise while developing an HPI predictor. 1) How should we select negative samples? 2) What should be the size of negative samples as compared to the positive samples? 3) What type of margin violation penalty should be used to train the predictor? We compare two available methods for negative sampling. Moreover, we propose a new method of assigning weights to each training example in weighted SVM depending on the distance of the negative examples from the positive examples. We have also developed a web server for our HPI predictor called HoPItor (Host Pathogen Interaction predicTOR) that can predict interactions between human and viral proteins. This webserver can be accessed at the URL: this http URL.

Title: A Useful Solution of the Coupon Collector's Problem, Abstract: The Coupon Collector's Problem is one of the few mathematical problems that make news headlines regularly. The reasons for this are on one hand the immense popularity of soccer albums (called Paninimania) and on the other hand that no solution is known that is able to take into account all effects such as replacement (limited purchasing of missing stickers) or swapping. In previous papers we have proven that the classical assumptions are not fulfilled in practice. Therefore we define new assumptions that match reality. Based on these assumptions we are able to derive formulae for the mean number of stickers needed (and the associated standard deviation) that are able to take into account all effects that occur in practical collecting. Thus collectors can estimate the average cost of completion of an album and its standard deviation just based on elementary calculations. From a practical point of view we consider the Coupon Collector's problem as solved. ----- Das Sammelbilderproblem ist eines der wenigen mathematischen Probleme, die regelmä{\ss}ig in den Schlagzeilen der Nachrichten vorkommen. Dies liegt einerseits an der gro{\ss}en Popularität von Fu{\ss}ball-Sammelbildern (Paninimania genannt) und andererseits daran, dass es bisher keine Lösung gibt, die alle relevanten Effekte wie Nachkaufen oder Tauschen berücksichtigt. Wir haben bereits nachgewiesen, dass die klassischen Annahmen nicht der Realität entsprechen. Deshalb stellen wir neue Annahmen auf, die die Praxis besser abbilden. Darauf aufbauend können wir Formeln für die mittlere Anzahl benötigter Bilder (sowie deren Standardabweichung) ableiten, die alle in der Praxis relevanten Effekte berücksichtigen. Damit können Sammler die mittleren Kosten eines Albums sowie deren Standardabweichung nur mit Hilfe von elementaren Rechnungen bestimmen. Für praktische Zwecke ist das Sammelbilderproblem damit gelöst.

Title: Accretion driven turbulence in filaments I: Non-gravitational accretion, Abstract: We study accretion driven turbulence for different inflow velocities in star forming filaments using the code ramses. Filaments are rarely isolated objects and their gravitational potential will lead to radially dominated accretion. In the non-gravitational case, accretion by itself can already provoke non-isotropic, radially dominated turbulent motions responsible for the complex structure and non-thermal line widths observed in filaments. We find that there is a direct linear relation between the absolute value of the total density weighted velocity dispersion and the infall velocity. The turbulent velocity dispersion in the filaments is independent of sound speed or any net flow along the filament. We show that the density weighted velocity dispersion acts as an additional pressure term supporting the filament in hydrostatic equilibrium. Comparing to observations, we find that the projected non-thermal line width variation is generally subsonic independent of inflow velocity.

Title: Contextual Outlier Interpretation, Abstract: Outlier detection plays an essential role in many data-driven applications to identify isolated instances that are different from the majority. While many statistical learning and data mining techniques have been used for developing more effective outlier detection algorithms, the interpretation of detected outliers does not receive much attention. Interpretation is becoming increasingly important to help people trust and evaluate the developed models through providing intrinsic reasons why the certain outliers are chosen. It is difficult, if not impossible, to simply apply feature selection for explaining outliers due to the distinct characteristics of various detection models, complicated structures of data in certain applications, and imbalanced distribution of outliers and normal instances. In addition, the role of contrastive contexts where outliers locate, as well as the relation between outliers and contexts, are usually overlooked in interpretation. To tackle the issues above, in this paper, we propose a novel Contextual Outlier INterpretation (COIN) method to explain the abnormality of existing outliers spotted by detectors. The interpretability for an outlier is achieved from three aspects: outlierness score, attributes that contribute to the abnormality, and contextual description of its neighborhoods. Experimental results on various types of datasets demonstrate the flexibility and effectiveness of the proposed framework compared with existing interpretation approaches.

Title: Game Theory for Secure Critical Interdependent Gas-Power-Water Infrastructure, Abstract: A city's critical infrastructure such as gas, water, and power systems, are largely interdependent since they share energy, computing, and communication resources. This, in turn, makes it challenging to endow them with fool-proof security solutions. In this paper, a unified model for interdependent gas-power-water infrastructure is presented and the security of this model is studied using a novel game-theoretic framework. In particular, a zero-sum noncooperative game is formulated between a malicious attacker who seeks to simultaneously alter the states of the gas-power-water critical infrastructure to increase the power generation cost and a defender who allocates communication resources over its attack detection filters in local areas to monitor the infrastructure. At the mixed strategy Nash equilibrium of this game, numerical results show that the expected power generation cost deviation is 35\% lower than the one resulting from an equal allocation of resources over the local filters. The results also show that, at equilibrium, the interdependence of the power system on the natural gas and water systems can motivate the attacker to target the states of the water and natural gas systems to change the operational states of the power grid. Conversely, the defender allocates a portion of its resources to the water and natural gas states of the interdependent system to protect the grid from state deviations.

Title: Implicit Entity Linking in Tweets, Abstract: Over the years, Twitter has become one of the largest communication platforms providing key data to various applications such as brand monitoring, trend detection, among others. Entity linking is one of the major tasks in natural language understanding from tweets and it associates entity mentions in text to corresponding entries in knowledge bases in order to provide unambiguous interpretation and additional con- text. State-of-the-art techniques have focused on linking explicitly mentioned entities in tweets with reasonable success. However, we argue that in addition to explicit mentions i.e. The movie Gravity was more ex- pensive than the mars orbiter mission entities (movie Gravity) can also be mentioned implicitly i.e. This new space movie is crazy. you must watch it!. This paper introduces the problem of implicit entity linking in tweets. We propose an approach that models the entities by exploiting their factual and contextual knowledge. We demonstrate how to use these models to perform implicit entity linking on a ground truth dataset with 397 tweets from two domains, namely, Movie and Book. Specifically, we show: 1) the importance of linking implicit entities and its value addition to the standard entity linking task, and 2) the importance of exploiting contextual knowledge associated with an entity for linking their implicit mentions. We also make the ground truth dataset publicly available to foster the research in this new research area.

Title: Non-cocompact Group Actions and $π_1$-Semistability at Infinity, Abstract: A finitely presented 1-ended group $G$ has {\it semistable fundamental group at infinity} if $G$ acts geometrically on a simply connected and locally compact ANR $Y$ having the property that any two proper rays in $Y$ are properly homotopic. This property of $Y$ captures a notion of connectivity at infinity stronger than "1-ended", and is in fact a feature of $G$, being independent of choices. It is a fundamental property in the homotopical study of finitely presented groups. While many important classes of groups have been shown to have semistable fundamental group at infinity, the question of whether every $G$ has this property has been a recognized open question for nearly forty years. In this paper we attack the problem by considering a proper {\it but non-cocompact} action of a group $J$ on such an $Y$. This $J$ would typically be a subgroup of infinite index in the geometrically acting over-group $G$; for example $J$ might be infinite cyclic or some other subgroup whose semistability properties are known. We divide the semistability property of $G$ into a $J$-part and a "perpendicular to $J$" part, and we analyze how these two parts fit together. Among other things, this analysis leads to a proof (in a companion paper) that a class of groups previously considered to be likely counter examples do in fact have the semistability property.

Title: Robust Distributed Control of DC Microgrids with Time-Varying Power Sharing, Abstract: This paper addresses the problem of output voltage regulation for multiple DC/DC converters connected to a microgrid, and prescribes a scheme for sharing power among different sources. This architecture is structured in such a way that it admits quantifiable analysis of the closed-loop performance of the network of converters; the analysis simplifies to studying closed-loop performance of an equivalent {\em single-converter} system. The proposed architecture allows for the proportion in which the sources provide power to vary with time; thus overcoming limitations of our previous designs. Additionally, the proposed control framework is suitable to both centralized and decentralized implementations, i.e., the same control architecture can be employed for voltage regulation irrespective of the availability of common load-current (or power) measurement, without the need to modify controller parameters. The performance becomes quantifiably better with better communication of the demanded load to all the controllers at all the converters (in the centralized case); however guarantees viability when such communication is absent. Case studies comprising of battery, PV and generic sources are presented and demonstrate the enhanced performance of prescribed optimal controllers for voltage regulation and power sharing.

Title: Quantum Lower Bounds for Tripartite Versions of the Hidden Shift and the Set Equality Problems, Abstract: In this paper, we study quantum query complexity of the following rather natural tripartite generalisations (in the spirit of the 3-sum problem) of the hidden shift and the set equality problems, which we call the 3-shift-sum and the 3-matching-sum problems. The 3-shift-sum problem is as follows: given a table of $3\times n$ elements, is it possible to circularly shift its rows so that the sum of the elements in each column becomes zero? It is promised that, if this is not the case, then no 3 elements in the table sum up to zero. The 3-matching-sum problem is defined similarly, but it is allowed to arbitrarily permute elements within each row. For these problems, we prove lower bounds of $\Omega(n^{1/3})$ and $\Omega(\sqrt n)$, respectively. The second lower bound is tight. The lower bounds are proven by a novel application of the dual learning graph framework and by using representation-theoretic tools.

Title: A Parallel Direct Cut Algorithm for High-Order Overset Methods with Application to a Spinning Golf Ball, Abstract: Overset methods are commonly employed to enable the effective simulation of problems involving complex geometries and moving objects such as rotorcraft. This paper presents a novel overset domain connectivity algorithm based upon the direct cut approach suitable for use with GPU-accelerated solvers on high-order curved grids. In contrast to previous methods it is capable of exploiting the highly data-parallel nature of modern accelerators. Further, the approach is also substantially more efficient at handling the curved grids which arise within the context of high-order methods. An implementation of this new algorithm is presented and combined with a high-order fluid dynamics code. The algorithm is validated against several benchmark problems, including flow over a spinning golf ball at a Reynolds number of 150,000.

Title: A data assimilation algorithm: the paradigm of the 3D Leray-alpha model of turbulence, Abstract: In this paper we survey the various implementations of a new data assimilation (downscaling) algorithm based on spatial coarse mesh measurements. As a paradigm, we demonstrate the application of this algorithm to the 3D Leray-$\alpha$ subgrid scale turbulence model. Most importantly, we use this paradigm to show that it is not always necessary that one has to collect coarse mesh measurements of all the state variables, that are involved in the underlying evolutionary system, in order to recover the corresponding exact reference solution. Specifically, we show that in the case of the 3D Leray$-\alpha$ model of turbulence the solutions of the algorithm, constructed using only coarse mesh observations of any two components of the three-dimensional velocity field, and without any information of the third component, converge, at an exponential rate in time, to the corresponding exact reference solution of the 3D Leray$-\alpha$ model. This study serves as an addendum to our recent work on abridged continuous data assimilation for the 2D Navier-Stokes equations. Notably, similar results have also been recently established for the 3D viscous Planetary Geostrophic circulation model in which we show that coarse mesh measurements of the temperature alone are sufficient for recovering, through our data assimilation algorithm, the full solution; viz. the three components of velocity vector field and the temperature. Consequently, this proves the Charney conjecture for the 3D Planetary Geostrophic model; namely, that the history of the large spatial scales of temperature is sufficient for determining all the other quantities (state variables) of the model.

Title: Playing a true Parrondo's game with a three state coin on a quantum walk, Abstract: Playing a Parrondo's game with a qutrit is the subject of this paper. We show that a true quantum Parrondo's game can be played with a 3 state coin(qutrit) in a 1D quantum walk in contrast to the fact that playing a true Parrondo's game with a 2 state coin(qubit) in 1D quantum walk fails in the asymptotic limits.

Title: Cross-modal Recurrent Models for Weight Objective Prediction from Multimodal Time-series Data, Abstract: We analyse multimodal time-series data corresponding to weight, sleep and steps measurements. We focus on predicting whether a user will successfully achieve his/her weight objective. For this, we design several deep long short-term memory (LSTM) architectures, including a novel cross-modal LSTM (X-LSTM), and demonstrate their superiority over baseline approaches. The X-LSTM improves parameter efficiency by processing each modality separately and allowing for information flow between them by way of recurrent cross-connections. We present a general hyperparameter optimisation technique for X-LSTMs, which allows us to significantly improve on the LSTM and a prior state-of-the-art cross-modal approach, using a comparable number of parameters. Finally, we visualise the model's predictions, revealing implications about latent variables in this task.

Title: Spatial Variational Auto-Encoding via Matrix-Variate Normal Distributions, Abstract: The key idea of variational auto-encoders (VAEs) resembles that of traditional auto-encoder models in which spatial information is supposed to be explicitly encoded in the latent space. However, the latent variables in VAEs are vectors, which can be interpreted as multiple feature maps of size 1x1. Such representations can only convey spatial information implicitly when coupled with powerful decoders. In this work, we propose spatial VAEs that use feature maps of larger size as latent variables to explicitly capture spatial information. This is achieved by allowing the latent variables to be sampled from matrix-variate normal (MVN) distributions whose parameters are computed from the encoder network. To increase dependencies among locations on latent feature maps and reduce the number of parameters, we further propose spatial VAEs via low-rank MVN distributions. Experimental results show that the proposed spatial VAEs outperform original VAEs in capturing rich structural and spatial information.

Title: Optimal Ramp Schemes and Related Combinatorial Objects, Abstract: In 1996, Jackson and Martin proved that a strong ideal ramp scheme is equivalent to an orthogonal array. However, there was no good characterization of ideal ramp schemes that are not strong. Here we show the equivalence of ideal ramp schemes to a new variant of orthogonal arrays that we term augmented orthogonal arrays. We give some constructions for these new kinds of arrays, and, as a consequence, we also provide parameter situations where ideal ramp schemes exist but strong ideal ramp schemes do not exist.

Title: Do Neural Nets Learn Statistical Laws behind Natural Language?, Abstract: The performance of deep learning in natural language processing has been spectacular, but the reasons for this success remain unclear because of the inherent complexity of deep learning. This paper provides empirical evidence of its effectiveness and of a limitation of neural networks for language engineering. Precisely, we demonstrate that a neural language model based on long short-term memory (LSTM) effectively reproduces Zipf's law and Heaps' law, two representative statistical properties underlying natural language. We discuss the quality of reproducibility and the emergence of Zipf's law and Heaps' law as training progresses. We also point out that the neural language model has a limitation in reproducing long-range correlation, another statistical property of natural language. This understanding could provide a direction for improving the architectures of neural networks.

Title: Super-speeds with Zero-RAM: Next Generation Large-Scale Optimization in Your Laptop!, Abstract: This article presents the novel breakthrough general purpose algorithm for large scale optimization problems. The novel algorithm is capable of achieving breakthrough speeds for very large-scale optimization on general purpose laptops and embedded systems. Application of the algorithm to the Griewank function was possible in up to 1 billion decision variables in double precision took only 64485 seconds (~18 hours) to solve, while consuming 7,630 MB (7.6 GB) or RAM on a single threaded laptop CPU. It shows that the algorithm is computationally and memory (space) linearly efficient, and can find the optimal or near-optimal solution in a fraction of the time and memory that many conventional algorithms require. It is envisaged that this will open up new possibilities of real-time large-scale problems on personal laptops and embedded systems.

Title: Recoverable Energy of Dissipative Electromagnetic Systems, Abstract: Ambiguities in the definition of stored energy within distributed or radiating electromagnetic systems motivate the discussion of the well-defined concept of recoverable energy. This concept is commonly overlooked by the community and the purpose of this communication is to recall its existence and to discuss its relationship to fractional bandwidth. Using a rational function approximation of a system's input impedance, the recoverable energy of lumped and radiating systems is calculated in closed form and is related to stored energy and fractional bandwidth. Lumped circuits are also used to demonstrate the relationship between recoverable energy and the energy stored within equivalent circuits produced by the minimum phase-shift Darlington's synthesis procedure.

Title: Elliptic Hall algebra on $\mathbb{F}_1$, Abstract: We construct Hall algebra of elliptic curve over $\mathbb{F}_1$ using the theory of monoidal scheme due to Deitmar and the theory of Hall algebra for monoidal representations due to Szczesny. The resulting algebra is shown to be a specialization of elliptic Hall algebra studied by Burban and Schiffmann. Thus our algebra is isomorphic to the skein algebra for torus by the recent work of Morton and Samuelson.

Title: Approximate Bayesian inference with queueing networks and coupled jump processes, Abstract: Queueing networks are systems of theoretical interest that give rise to complex families of stochastic processes, and find widespread use in the performance evaluation of interconnected resources. Yet, despite their importance within applications, and in comparison to their counterpart stochastic models in genetics or mathematical biology, there exist few relevant approaches for transient inference and uncertainty quantification tasks in these systems. This is a consequence of strong computational impediments and distinctive properties of the Markov jump processes induced by queueing networks. In this paper, we offer a comprehensive overview of the inferential challenge and its comparison to analogue tasks within related mathematical domains. We then discuss a model augmentation over an approximating network system, and present a flexible and scalable variational Bayesian framework, which is targeted at general-form open and closed queueing systems, with varied service disciplines and priorities. The inferential procedure is finally validated in a couple of uncertainty quantification tasks for network service rates.

Title: Universal features of price formation in financial markets: perspectives from Deep Learning, Abstract: Using a large-scale Deep Learning approach applied to a high-frequency database containing billions of electronic market quotes and transactions for US equities, we uncover nonparametric evidence for the existence of a universal and stationary price formation mechanism relating the dynamics of supply and demand for a stock, as revealed through the order book, to subsequent variations in its market price. We assess the model by testing its out-of-sample predictions for the direction of price moves given the history of price and order flow, across a wide range of stocks and time periods. The universal price formation model is shown to exhibit a remarkably stable out-of-sample prediction accuracy across time, for a wide range of stocks from different sectors. Interestingly, these results also hold for stocks which are not part of the training sample, showing that the relations captured by the model are universal and not asset-specific. The universal model --- trained on data from all stocks --- outperforms, in terms of out-of-sample prediction accuracy, asset-specific linear and nonlinear models trained on time series of any given stock, showing that the universal nature of price formation weighs in favour of pooling together financial data from various stocks, rather than designing asset- or sector-specific models as commonly done. Standard data normalizations based on volatility, price level or average spread, or partitioning the training data into sectors or categories such as large/small tick stocks, do not improve training results. On the other hand, inclusion of price and order flow history over many past observations is shown to improve forecasting performance, showing evidence of path-dependence in price dynamics.

Title: A New Tracking Algorithm for Multiple Colloidal Particles Close to Contact, Abstract: In this paper, we propose a new algorithm based on radial symmetry center method to track colloidal particles close to contact, where the optical images of the particles start to overlap in digital video microscopy. This overlapping effect is important to observe the pair interaction potential in colloidal studies and it appears as additional interaction in the measurement of the interaction with conventional tracking analysis. The proposed algorithm in this work is simple, fast and applicable for not only two particles but also three and more particles without any modification. The algorithm uses gradient vectors of the particle intensity distribution, which allows us to use a part of the symmetric intensity distribution in the calculation of the actual particle position. In this study, simulations are performed to see the performance of the proposed algorithm for two and three particles, where the simulation images are generated by using fitted curve to experimental particle image for different sized particles. As a result, the algorithm yields the maximum error smaller than 2 nm for 5.53 {\mu}m silica particles in contact condition.

Title: Critical Percolation Without Fine Tuning on the Surface of a Topological Superconductor, Abstract: We present numerical evidence that most two-dimensional surface states of a bulk topological superconductor (TSC) sit at an integer quantum Hall plateau transition. We study TSC surface states in class CI with quenched disorder. Low-energy (finite-energy) surface states were expected to be critically delocalized (Anderson localized). We confirm the low-energy picture, but find instead that finite-energy states are also delocalized, with universal statistics that are independent of the TSC winding number, and consistent with the spin quantum Hall plateau transition (percolation).

Title: Low-luminosity stellar wind accretion onto neutron stars in HMXBs, Abstract: Features and applications of quasi-spherical settling accretion onto rotating magnetized neutron stars in high-mass X-ray binaries are discussed. The settling accretion occurs in wind-fed HMXBs when the plasma cooling time is longer than the free-fall time from the gravitational capture radius, which can take place in low-luminosity HMXBs with $L_x\lesssim 4\times 10^{36}$ erg/s. We briefly review the implications of the settling accretion, focusing on the SFXT phenomenon, which can be related to instability of the quasi-spherical convective shell above the neutron star magnetosphere due to magnetic reconnection from fast temporarily magnetized winds from OB-supergiant. If a young neutron star in a wind-fed HMXB is rapidly rotating, the propeller regime in a quasi-spherical hot shell occurs. We show that X-ray spectral and temporal properties of enigmatic $\gamma$ Cas Be-stars are consistent with failed settling accretion regime onto a propelling neutron star. The subsequent evolutionary stage of $\gamma$ Cas and its analogs should be the X Per-type binaries comprising low-luminosity slowly rotating X-ray pulsars.

Title: Faithful Inversion of Generative Models for Effective Amortized Inference, Abstract: Inference amortization methods share information across multiple posterior-inference problems, allowing each to be carried out more efficiently. Generally, they require the inversion of the dependency structure in the generative model, as the modeller must learn a mapping from observations to distributions approximating the posterior. Previous approaches have involved inverting the dependency structure in a heuristic way that fails to capture these dependencies correctly, thereby limiting the achievable accuracy of the resulting approximations. We introduce an algorithm for faithfully, and minimally, inverting the graphical model structure of any generative model. Such inverses have two crucial properties: (a) they do not encode any independence assertions that are absent from the model and; (b) they are local maxima for the number of true independencies encoded. We prove the correctness of our approach and empirically show that the resulting minimally faithful inverses lead to better inference amortization than existing heuristic approaches.

Title: A social Network Analysis of the Operations Research/Industrial Engineering Faculty Hiring Network, Abstract: We study the U.S. Operations Research/Industrial-Systems Engineering (ORIE) faculty hiring network, consisting of 1,179 faculty origin and destination data together with attribute data from 83 ORIE departments. A social network analysis of faculty hires can reveal important patterns in an academic field, such as the existence of a hierarchy or sociological aspects such as the presence of communities of departments. We first statistically test for the existence of a linear hierarchy in the network and for its steepness. We find a near linear hierarchical order of the departments, proposing a new index for hiring networks, which we contrast with other indicators of hierarchy, including published rankings. A single index is not capable to capture the full structure of a complex network, however, so we next fit a latent exponential random graph model (ERGM) to the network, which is able to reproduce its main observed characteristics: high incidence of self-hiring, skewed out-degree distribution, low density and clustering. Finally, we use the latent variables in the ERGM to simplify the network to one where faculty hires take place among three groups of departments. We contrast our findings with those reported for other related disciplines, Computer Science and Business.

Title: One-sample aggregate data meta-analysis of medians, Abstract: An aggregate data meta-analysis is a statistical method that pools the summary statistics of several selected studies to estimate the outcome of interest. When considering a continuous outcome, typically each study must report the same measure of the outcome variable and its spread (e.g., the sample mean and its standard error). However, some studies may instead report the median along with various measures of spread. Recently, the task of incorporating medians in meta-analysis has been achieved by estimating the sample mean and its standard error from each study that reports a median in order to meta-analyze the means. In this paper, we propose two alternative approaches to meta-analyze data that instead rely on medians. We systematically compare these approaches via simulation study to each other and to methods that transform the study-specific medians and spread into sample means and their standard errors. We demonstrate that the proposed median-based approaches perform better than the transformation-based approaches, especially when applied to skewed data and data with high inter-study variance. In addition, when meta-analyzing data that consists of medians, we show that the median-based approaches perform considerably better than or comparably to the best-case scenario for a transformation approach: conducting a meta-analysis using the actual sample mean and standard error of the mean of each study. Finally, we illustrate these approaches in a meta-analysis of patient delay in tuberculosis diagnosis.

Title: QuickCast: Fast and Efficient Inter-Datacenter Transfers using Forwarding Tree Cohorts, Abstract: Large inter-datacenter transfers are crucial for cloud service efficiency and are increasingly used by organizations that have dedicated wide area networks between datacenters. A recent work uses multicast forwarding trees to reduce the bandwidth needs and improve completion times of point-to-multipoint transfers. Using a single forwarding tree per transfer, however, leads to poor performance because the slowest receiver dictates the completion time for all receivers. Using multiple forwarding trees per transfer alleviates this concern--the average receiver could finish early; however, if done naively, bandwidth usage would also increase and it is apriori unclear how best to partition receivers, how to construct the multiple trees and how to determine the rate and schedule of flows on these trees. This paper presents QuickCast, a first solution to these problems. Using simulations on real-world network topologies, we see that QuickCast can speed up the average receiver's completion time by as much as $10\times$ while only using $1.04\times$ more bandwidth; further, the completion time for all receivers also improves by as much as $1.6\times$ faster at high loads.

Title: Contemporary machine learning: a guide for practitioners in the physical sciences, Abstract: Machine learning is finding increasingly broad application in the physical sciences. This most often involves building a model relationship between a dependent, measurable output and an associated set of controllable, but complicated, independent inputs. We present a tutorial on current techniques in machine learning -- a jumping-off point for interested researchers to advance their work. We focus on deep neural networks with an emphasis on demystifying deep learning. We begin with background ideas in machine learning and some example applications from current research in plasma physics. We discuss supervised learning techniques for modeling complicated functions, beginning with familiar regression schemes, then advancing to more sophisticated deep learning methods. We also address unsupervised learning and techniques for reducing the dimensionality of input spaces. Along the way, we describe methods for practitioners to help ensure that their models generalize from their training data to as-yet-unseen test data. We describe classes of tasks -- predicting scalars, handling images, fitting time-series -- and prepare the reader to choose an appropriate technique. We finally point out some limitations to modern machine learning and speculate on some ways that practitioners from the physical sciences may be particularly suited to help.

Title: Uniform diamond coatings on WC-Co hard alloy cutting inserts deposited by a microwave plasma CVD, Abstract: Polycrystalline diamond coatings have been grown on cemented carbide substrates with different aspect ratios by a microwave plasma CVD in methane-hydrogen gas mixtures. To protect the edges of the substrates from non-uniform heating due to the plasma edge effect, a special plateholder with pockets for group growth has been used. The difference in heights of the substrates and plateholder, and its influence on the diamond film mean grain size, growth rate, phase composition and stress was investigated. The substrate temperature range, within which uniform diamond films are produced with good adhesion, is determined. The diamond-coated cutting inserts produced at optimized process exhibited a reduction of cutting force and wear resistance by a factor of two, and cutting efficiency increase by 4.3 times upon turning A390 Al-Si alloy as compared to performance of uncoated tools.

Title: Analysing Soccer Games with Clustering and Conceptors, Abstract: We present a new approach for identifying situations and behaviours, which we call "moves", from soccer games in the 2D simulation league. Being able to identify key situations and behaviours are useful capabilities for analysing soccer matches, anticipating opponent behaviours to aid selection of appropriate tactics, and also as a prerequisite for automatic learning of behaviours and policies. To support a wide set of strategies, our goal is to identify situations from data, in an unsupervised way without making use of pre-defined soccer specific concepts such as "pass" or "dribble". The recurrent neural networks we use in our approach act as a high-dimensional projection of the recent history of a situation on the field. Similar situations, i.e., with similar histories, are found by clustering of network states. The same networks are also used to learn so-called conceptors, that are lower-dimensional manifolds that describe trajectories through a high-dimensional state space that enable situation-specific predictions from the same neural network. With the proposed approach, we can segment games into sequences of situations that are learnt in an unsupervised way, and learn conceptors that are useful for the prediction of the near future of the respective situation.

Title: On the Efficient Simulation of the Left-Tail of the Sum of Correlated Log-normal Variates, Abstract: The sum of Log-normal variates is encountered in many challenging applications such as in performance analysis of wireless communication systems and in financial engineering. Several approximation methods have been developed in the literature, the accuracy of which is not ensured in the tail regions. These regions are of primordial interest wherein small probability values have to be evaluated with high precision. Variance reduction techniques are known to yield accurate, yet efficient, estimates of small probability values. Most of the existing approaches, however, have considered the problem of estimating the right-tail of the sum of Log-normal random variables (RVS). In the present work, we consider instead the estimation of the left-tail of the sum of correlated Log-normal variates with Gaussian copula under a mild assumption on the covariance matrix. We propose an estimator combining an existing mean-shifting importance sampling approach with a control variate technique. The main result is that the proposed estimator has an asymptotically vanishing relative error which represents a major finding in the context of the left-tail simulation of the sum of Log-normal RVs. Finally, we assess by various simulation results the performances of the proposed estimator compared to existing estimators.

Title: Why optional stopping is a problem for Bayesians, Abstract: Recently, optional stopping has been a subject of debate in the Bayesian psychology community. Rouder (2014) argues that optional stopping is no problem for Bayesians, and even recommends the use of optional stopping in practice, as do Wagenmakers et al. (2012). This article addresses the question whether optional stopping is problematic for Bayesian methods, and specifies under which circumstances and in which sense it is and is not. By slightly varying and extending Rouder's (2014) experiment, we illustrate that, as soon as the parameters of interest are equipped with default or pragmatic priors - which means, in most practical applications of Bayes Factor hypothesis testing - resilience to optional stopping can break down. We distinguish between four types of default priors, each having their own specific issues with optional stopping, ranging from no-problem-at-all (Type 0 priors) to quite severe (Type II and III priors).