Chapter 11 – Game playing

11.1  Overview

11.2  Introduction

11.3  Characteristics of Game Playing

11.4  Standard Games

11.4.1  A Simple Game Tree

11.4.2  Heuristics and Minimax Search

11.4.3  Horizon Problems

11.4.4  Alpha--beta Pruning

11.4.5  The Imperfect Opponent

11.5  Non-zero-sum Games and Simultaneous Play

11.5.1  The Prisoner's Dilemma

11.5.2  Searching the Game Tree

11.5.3  No Alpha--Beta Pruning

11.5.4  Pareto-optimality

11.5.5  Multi-party Competition and Co-operation

11.6  The Adversary Is Life!

11.7  Probability

11.8  Neural Networks for Games

11.8.1  Where to Use a Neural Network

11.8.2  Training Data and Self Play

11.9  Summary

adversarial learning, agents, alpha–beta pruning, AlphaGo, AlphaGo Zero, Analytical Engine, Babbage, Charles, bootstrapping, branch and bound search, branching factor, breadth first search, chess heuristic, chess program, coin-weighing problem, computer chess, Cuban missile crisis, deep neural network, deterministic search, distributed AI, dominoes, game playing!heuristics, game theory, game tree, genetic algorithm, Go, heuristic evaluation function, hill climbing algorithm, human intelligence, iterative deepening, knowledge-rich search, Lee Sedol, machine learning, magic square, minimax score, minimax search, Monte Carlo tree search, neural network, non-zero-sum game, noughts and crosses, optimal solution, Pareto-optimal, pattern matching, placing dominoes, plateau, policy network, prisoner's dilemma, probabilistic reasoning, probability, probability-based cut-off, risk avoidance, robotics, search horizon, search space, search strategies, search tree, self play, stochastic search, uncertainty, value network, zero-sum game