266 Hits in 7.3 sec

Improving heuristic mini-max search by supervised learning

Michael Buro
2002 Artificial Intelligence  
This article surveys three techniques for enhancing heuristic game-tree search pioneered in the author's Othello program LOGISTELLO, which dominated the computer Othello scene for several years and won  ...  These general methods represent the state-of-the-art in computer Othello programming and begin to attract researchers in related fields.  ...  Evaluation function learning Many AI systems use evaluation functions for guiding search tasks.  ... 
doi:10.1016/s0004-3702(01)00093-5 fatcat:vzfhumuzyngrjihjr2mdbdhdsm

OLIVAW: Mastering Othello without Human Knowledge, nor a Fortune [article]

Antonio Norelli, Alessandro Panconesi
2022 arXiv   pre-print
While being simpler than Chess or Go, Othello maintains a considerable search space and difficulty in evaluating board positions.  ...  In this paper, we show how the AlphaGo Zero's paradigm can be successfully applied to the popular game of Othello using only commodity hardware and free cloud services.  ...  OLIVAW's style of play, Roberto Sperandio for the gripping match commentary, Benedetto Romano for the discussion on traditional engines, Leonardo Caviola for having introduced the public to Othello during  ... 
arXiv:2103.17228v4 fatcat:ucttbbxrdfct7c44yx6ilnus7e


Jonathan Baxter, Andrew Tridgell, Lex Weaver
2012 Machine Learning  
We present some experiments in which our chess program "KnightCap" used TDLEAF( ) to learn its evaluation function while playing on Internet chess servers.  ...  We discuss some of the reasons for this success, principle among them being the use of on-line, rather than self-play.  ...  Unfortunately, for games like othello and chess it is very difficult to accurately evaluate a position by looking only one move or ply ahead.  ... 
doi:10.1023/a:1007634325138 fatcat:6oelfeyzsbdjhh6bqtfakqrsp4

Learning n-tuple networks for othello by coevolutionary gradient search

Krzysztof Krawiec, Marcin Grzegorz Szubert
2011 Proceedings of the 13th annual conference on Genetic and evolutionary computation - GECCO '11  
The approach is applied to learning Othello strategies represented as n-tuple networks using different search operators and modes of learning.  ...  We propose Coevolutionary Gradient Search, a blueprint for a family of iterative learning algorithms that combine elements of local search and population-based search.  ...  They often incorporate supervised learning techniques that use large expertlabeled game databases and efficient look-ahead game tree search.  ... 
doi:10.1145/2001576.2001626 dblp:conf/gecco/KrawiecS11 fatcat:mky7oxm2ijfirbnulz6jd3dptm

KnightCap: A chess program that learns by combining TD(lambda) with game-tree search [article]

Jonathan Baxter, Andrew Tridgell, Lex Weaver
1999 arXiv   pre-print
We present some experiments in which our chess program "KnightCap" used TDLeaf(lambda) to learn its evaluation function while playing on the Free Internet Chess Server (FICS,  ...  We discuss some of the reasons for this success, principle among them being the use of on-line, rather than self-play.  ...  Acknowledgements Thanks to several of the anonymous referees for their helpful remarks. Jonathan Baxter was supported by an Australian Postdoctoral Fellowship.  ... 
arXiv:cs/9901002v1 fatcat:4oinsgk6hvaihi37mggtpfpar4

TDLeaf(lambda): Combining Temporal Difference Learning with Game-Tree Search [article]

Jonathan Baxter, Andrew Tridgell, Lex Weaver
1999 arXiv   pre-print
In particular, our chess program, "KnightCap," used TDLeaf(lambda) to learn its evaluation function while playing on the Free Internet Chess Server (FICS,  ...  In this paper we present TDLeaf(lambda), a variation on the TD(lambda) algorithm that enables it to be used in conjunction with minimax search.  ...  Unfortunately, for games like othello and chess it is difficult to accurately evaluate a position by looking only one move or ply ahead.  ... 
arXiv:cs/9901001v1 fatcat:7bwqkpldjrcmfjxrkqbhslgsda

Coevolutionary Temporal Difference Learning for Othello

Marcin Szubert, Wojciech Jaskowski, Krzysztof Krawiec
2009 2009 IEEE Symposium on Computational Intelligence and Games  
We apply CTDL to the board game of Othello, using weighted piece counter for representing players' strategies.  ...  The coevolutionary part of the algorithm provides for exploration of the solution space, while the temporal difference learning performs its exploitation by local search.  ...  ACKNOWLEDGMENTS This work was supported in part by Ministry of Science and Higher Education grant # N N519 3505 33 and grant POIG.01.01.02-00-014/08-00.  ... 
doi:10.1109/cig.2009.5286486 dblp:conf/cig/SzubertJK09 fatcat:2byzeqgxzbb33ju7fhbekwlli4

AlphaZero-Inspired General Board Game Learning and Playing [article]

Johannes Scheiermann, Wolfgang Konen
2022 arXiv   pre-print
In this paper, we pick an important element of AlphaZero - the Monte Carlo Tree Search (MCTS) planning stage - and combine it with reinforcement learning (RL) agents.  ...  Recently, the seminal algorithms AlphaGo and AlphaZero have started a new era in game learning and deep reinforcement learning.  ...  TD-learning uses the value function V (s t ), which is the expected sum of future rewards when being in state s t .  ... 
arXiv:2204.13307v1 fatcat:v2u4b3pylnhknn3ros4nullxsm

Computing Games: Bridging The Gap Between Search and Entertainment

Anggina Primanita, Mohd Nor Akmal Khalid, Hiroyuki Iida
2021 IEEE Access  
In this case, PPNS is a better performing algorithm in terms of convergence for games or related areas that require a long look-ahead search.  ...  It needs a clear look-ahead tile placement strategy for the player to reach the desirable highest tile.  ...  He received his Ph.D. on heuristic theories of game tree search in 1994 from the Tokyo University of Agriculture and Technology, Tokyo.  ... 
doi:10.1109/access.2021.3079356 fatcat:gitthugluvhghgl5fqin22v4lu

Evolutionary computation and games

S.M. Lucas, G. Kendall
2006 IEEE Computational Intelligence Magazine  
Go strategy seems to rely as much on pattern recognition as it does on logical analysis, and the large branching factor severely restricts the look-ahead that can be used within a game-tree search.  ...  Learning techniques were being used for checkers as far back as the 1950s with Samuel's seminal work ([15], which was reproduced in [16] ).  ...  Acknowledgments We thank Thomas Runarsson for discussions related to this article.  ... 
doi:10.1109/mci.2006.1597057 fatcat:6o7yidxnirftvarbpx2fcxqh44

*-Minimax Performance in Backgammon [chapter]

Thomas Hauk, Michael Buro, Jonathan Schaeffer
2006 Lecture Notes in Computer Science  
We also present empirical evidence that with today's sophisticated evaluation functions good checker play in backgammon does not require deep searches.  ...  Star2 allows strong backgammon programs to conduct depth 5 full-width searches (up from 3) under tournament conditions on regular hardware without using risky forward pruning techniques.  ...  Acknowledgements This research was supported by grants from the Natural Sciences and Engineering Research Council of Canada (NSERC) and Alberta's Informatics Circle of Research Excellence (iCORE).  ... 
doi:10.1007/11674399_4 fatcat:cbfhyqxpr5bitcgbqdomoxdnf4

CadiaPlayer: A Simulation-Based General Game Player

Y. Bjornsson, H. Finnsson
2009 IEEE Transactions on Computational Intelligence and AI in Games  
The traditional design model for GGP agents has been to use a minimax-based game-tree search augmented with an automatically learned heuristic evaluation function.  ...  In here we describe CADIAPLAYER, a GGP agent employing a radically different approach: instead of a traditional game-tree search it uses Monte-Carlo simulations for its move decisions.  ...  Domain-dependent knowledge plays an important part in both components, in particular for game-position evaluation, understandably, but also for providing effective search guidance.  ... 
doi:10.1109/tciaig.2009.2018702 fatcat:2oe2227p7ngfjnngquvn22rekq

Reinforcement Learning with N-tuples on the Game Connect-4 [chapter]

Markus Thill, Patrick Koch, Wolfgang Konen
2012 Lecture Notes in Computer Science  
We apply temporal difference learning (TDL), a well-known variant of the reinforcement learning approach, in combination with n-tuple networks to the game Connect-4.  ...  Learning complex game functions is still a difficult task.  ...  Each n-tuple is allowed to have one or two LUTs (see Sec. 4). 1 During training and play, the TDL agent does not use any game-tree search (0-ply look-ahead), instead it just inspects the board positions  ... 
doi:10.1007/978-3-642-32937-1_19 fatcat:wg3xezzn7rbxjcajmajiqmlcx4

A Gamut of Games

Jonathan Schaeffer
2001 The AI Magazine  
This article reviews the past successes, current projects, and future research directions for AI using computer games as a research test bed. Articles  ...  In Shannon's time, it would have seemed unlikely that only a scant 50 years would be needed to develop programs that play world-class backgammon, checkers, chess, Othello, and Scrabble.  ...  an article on computer games in volume 1 of this series.  ... 
doi:10.1609/aimag.v22i3.1570 dblp:journals/aim/Schaeffer01 fatcat:6zirpq2v2va3bh7hkwhtaa577y

On Achieving History-Based Move Ordering in Adversarial Board Games Using Adaptive Data Structures [chapter]

Spencer Polk, B. John Oommen
2016 Lecture Notes in Computer Science  
This paper concerns the problem of enhancing the well-known alpha-beta search technique for intelligent game playing.  ...  These involve providing a bound on the size of the ADS, based on the hypothesis that it can retain most of its benefits with a smaller list, and examining the possibility of using a different ADS for each  ...  The game tree is explored in a depth-first manner, until the desired ply is reached, at which point the game state is evaluated according to some form of refined heuristic, assigning a value to the position  ... 
doi:10.1007/978-3-662-49619-0_2 fatcat:c4m3m537avfplksezzrdxk2sje
« Previous Showing results 1 — 15 out of 266 results