Neil Burch

TL;DR: This work describes an end-to-end approach for building an implicit modelling agent, compute robust response strategies, show how to select strategies for the portfolio, and apply existing variance reduction and online learning techniques to dynamically adapt the agent's strategy to its opponent.

TL;DR: A new type of database, the Local Distance Database (LDDB), that contains distances between boundary points of a local neighborhood that calculates the optimal path between start and goal locations given the local distances stored in the LDDB is introduced.

Abstract: AI has had notable success in building high-performance game-playing programs to complete against the best human players. However, the availability of fast and plentiful machines with large memories and disks creates the possibility of solving a game. This has been done before for simple or relatively small games. In this paper, we present new ideas and algorithms for solving the game of checkers. Checkers is a popular game of skill with a search space of 1020 possible positions. This paper reports on our first result. One of the most challenging checkers openings has been solved-the White Doctor opening is a draw. Solving roughly 50 more openings will result in the game-theoretic value of checkers being determined.

TL;DR: A new measure of game complexity that links existing state-of-the-art algorithms for computing approximate equilibria to a more human measure, which considers the range of skill in a game, i.e. how many different skill levels exist.

TL;DR: DeepNash as discussed by the authors uses a game-theoretic, model-free deep reinforcement learning method, without search, that learns to master Stratego through self-play from scratch.