Learning to Play the Game of Chess

By Sebastian Thrun. 1994

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Reinforcement Learning

Citations in the corpus

Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm (2017, neutral): “NeuroChess (33) evaluated positions by a neural network that used 175 handcrafted input features.”
Giraffe: Using Deep Reinforcement Learning to Play Chess (2015, methodology): “For example, Thrun reported a diﬀerence of 2 orders of magnitude between his implementation of a neural-network based chess engine and a state of art chess engine of that time [18].”
Deep Imitative Models for Flexible Inference, Planning, and Control (2018, methodology): “However, in practice, model-based and model-free RL algorithms are vulnerable to distributional drift (Thrun, 1995; Ross & Bagnell, 2010): when acting according to the learned model or policy, the agent visits states different from those seen during training, and in those it is unlikely to…”
Reinforcement Learning: A Survey (1996, background): “Other games that have been studied include Go (Schraudolph, Dayan, & Sejnowski, 1994) and Chess (Thrun, 1995).”
Safe, Multi-Agent, Reinforcement Learning for Autonomous Driving (2016, neutral): “See also [39, 38, 18, 8].”
Generalized Markov Decision Processes: Dynamic-programming and Reinforcement-learning Algorithms (1996, background): “The theory of Markov decision processes can be used as a theoretical foundation forimportant results concerning this decision-making problem [2].”
Robust Supervised Learning (2005, background): “(Thrun 1995, Pomerleau 1989) Consider the widely-known machine learning success of ALVINN, a neural network designed to drive at speed in both on-and off-road environments.”
TDLeaf(lambda): Combining Temporal Difference Learning with Game-Tree Search (1999, background): “In contrast, ﬁnding a representation for chess, othello or Go which allows a small neural network to order moves at one-ply with near human performance is a far more difﬁcult task [9, 11, 5].”
Algorithms for Sequential Decision Making (1996, background): “Othergames that have been studied include Go [139] and Chess [161].”
Adversarial Reinforcement Learning (1997, methodology): “[ Thrun, 1995] was moderately successful in using similar techniques in chess, but these techniques were not as successful as they had been in the checkers domain.”
Experiments in Parameter Learning Using Temporal Differences (1998, methodology): “We discuss the details of our learning algorithm, details of KnightCap, and some of the principal reasons for KnightCap’sr id improvement.”
Efficient Reductions for Imitation Learning (2010, background): “Ignoring the effects of the learner on the underlying state distribution leads to serious practical difﬁculties (Thrun 1995, Pomerleau 1989).”
Reinforcement learning in the game of Othello: Learning against a fixed opponent and learning from self-play (2013, background): “Other RL applications to games include chess [4], checkers [5] and Go [6].”
Cooperative Multi-Agent Learning: The State of the Art (2005, unknown): “Other games have included Tic-Tac-Toe [4], Mancala [61], Othello [241], Go [149], Chess [138, 269], Poker [139], Blackjack [261], Nim and 3D Tic-TacToe [219], and Tag [215].”
Deep learning in robotics: a review of recent research (2017, background): “In hindsight, we see that chess was considered in the early years of artificial intelligence to be representative of human intelligence over machines [24].”