background blrup and Methods section

app.jl

@@ -82,6 +82,17 @@ function ui(model)
       p([
          button("CESPF Greedy Rollout", @click("rollout = true"))
       ])
+      p([
+         h4("Background")
+        "Imperfect information games are games like Poker, or Uno, where information about some of the players' states are unknown."
+        " Conversely, Chess happens to be a perfect information game. Everyone knows where everyone's pieces are on the board. There's no hidden information."
+        " Dark Chess takes both these ideas and combines them, to make a really F**ing hard game to play, and even harder to model via an algorithm or machine learning problem. Not only are numerous actions you can take with your pieces,"
+      " You also have to extrapolate an exponentially growing set of possible states that your opponent is in. The search space ends up being very very large and intractible for traditional imperfect info (zero-sum) algorithms like CFR."
+      br()
+      h4("Methods Learned")
+      "In this interactive paper, we're going to explore the combination of two novel methods from Stephen McAleer, specifically Neural E(X)tensive-form Double Oracle, along with ESCHEW to build a top Dark Chess Engine."
+      "We also will explore ideas of Neural Replicator Dynamics and Follow the Regularized Leader to explore how we can optimize our Deep RL models to traverse policy gradients more effectively."
+      ])
     ],
     @iif(:isready)
   )