Filters








4,154 Hits in 6.1 sec

Mastering the game of Go with deep neural networks and tree search

David Silver, Aja Huang, Chris J. Maddison, Arthur Guez, Laurent Sifre, George van den Driessche, Julian Schrittwieser, Ioannis Antonoglou, Veda Panneershelvam, Marc Lanctot, Sander Dieleman, Dominik Grewe (+8 others)
2016 Nature  
Without any lookahead search, the neural networks play Go at the level of state-of-the-art Monte-Carlo tree search programs that simulate thousands of random games of self-play.  ...  These deep neural networks are trained by a novel combination of supervised learning from human expert games, and reinforcement learning from games of self-play.  ...  Acknowledgements We thank Fan Hui for agreeing to play against AlphaGo; Toby Manning for refereeing the match; Ostrovski for reviewing the paper; and the rest of the DeepMind team for their support, ideas  ... 
doi:10.1038/nature16961 pmid:26819042 fatcat:hhxixsirtjairjuwqivmi3gcga

Mastering the game of Go without human knowledge

David Silver, Julian Schrittwieser, Karen Simonyan, Ioannis Antonoglou, Aja Huang, Arthur Guez, Thomas Hubert, Lucas Baker, Matthew Lai, Adrian Bolton, Yutian Chen, Timothy Lillicrap (+5 others)
2017 Nature  
Recently, AlphaGo became the first program to defeat a world champion in the game of Go. The tree search in AlphaGo evaluated positions and selected moves using deep neural networks.  ...  This neural network improves the strength of tree search, resulting in higher quality move selection and stronger self-play in the next iteration.  ...  Cain for work on the visuals; A. Barreto, G. Ostrovski, T. Ewalds, T. Schaul, J. Oh and N. Heess for reviewing the paper; and the rest of the DeepMind team for their support.  ... 
doi:10.1038/nature24270 pmid:29052630 fatcat:h2n334a2ejfxtknx67kbiaswfq

Towards "AlphaChem": Chemical Synthesis Planning with Tree Search and Deep Neural Network Policies [article]

Marwin Segler, Mike Preuß, Mark P. Waller
2017 arXiv   pre-print
In combination with a Deep Neural Network policy learned from essentially the complete published knowledge of chemistry, Monte Carlo Tree Search (MCTS) can be used to evaluate positions.  ...  In exploratory studies, we demonstrate that MCTS with neural network policies outperforms the traditionally used best-first search with hand-coded heuristics.  ...  ACKNOWLEDGMENTS M.S. and M.P.W. thank RELX Intellectual Properties for the reaction dataset and Deutsche Forschungsgemeinschaft (SFB858) for funding.  ... 
arXiv:1702.00020v1 fatcat:f64ciznjmnezvh7qlqp3petm6i

人工知能,機械学習によるデータマイニング

石井 一夫
2016 Proceedings of the Annual Meeting of Japanese Society of Computational Statistics  
Mastering the game of Go with deep neural networks and tree search. Nature. 529(7587): 484-9(2016). 3) Ishii K, Kobayashi T, Toshinori K, Yamagata Y.  ...  /nature16961.Mastering the game of Go with deep neural networks and tree search.Silver D1, Huang A1, Maddison CJ1, Guez A, Sifre L, van den Driessche G, Schrittwieser J, Antonoglou I, Panneershelvam V,  ... 
doi:10.20551/jscstaikai.30.0_85 fatcat:7b7okxcn5zfancgoqpem3ozlle

A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play

David Silver, Thomas Hubert, Julian Schrittwieser, Ioannis Antonoglou, Matthew Lai, Arthur Guez, Marc Lanctot, Laurent Sifre, Dharshan Kumaran, Thore Graepel, Timothy Lillicrap, Karen Simonyan (+1 others)
2018 Science  
Starting from random play and given no domain knowledge except the game rules, AlphaZero convincingly defeated a world champion program in the games of chess and shogi (Japanese chess), as well as Go.  ...  By contrast, the AlphaGo Zero program recently achieved superhuman performance in the game of Go by reinforcement learning from self-play.  ...  Ostrovski for reviewing the paper; and the rest of the DeepMind team for their support. SUPPLEMENTARY MATERIALS www.sciencemag.org/content/362/6419/1140/suppl/DC1 Materials and Methods Figs.  ... 
doi:10.1126/science.aar6404 pmid:30523106 fatcat:3ojohsnggndppnfm5akucp7pve

Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm [article]

David Silver, Thomas Hubert, Julian Schrittwieser, Ioannis Antonoglou, Matthew Lai, Arthur Guez, Marc Lanctot, Laurent Sifre, Dharshan Kumaran, Thore Graepel, Timothy Lillicrap, Karen Simonyan, Demis Hassabis
2017 arXiv   pre-print
In contrast, the AlphaGo Zero program recently achieved superhuman performance in the game of Go, by tabula rasa reinforcement learning from games of self-play.  ...  Starting from random play, and given no domain knowledge except the game rules, AlphaZero achieved within 24 hours a superhuman level of play in the games of chess and shogi (Japanese chess) as well as  ...  Mastering the game of Go with deep neural networks and tree search. 30. Wilhelm Steinitz. The Modern Chess Instructor. Edition Olms AG, 1990. 31. Sebastian Thrun. Learning to play the game of chess.  ... 
arXiv:1712.01815v1 fatcat:flj56adezzf6xepdezevbo24xq

Lessons Learned from AlphaGo

Steffen Hölldobler, Sibylle Möhle, Anna Tigunova
2017 Young Scientists' International Workshop on Trends in Information Processing  
The system they designed is based on tree search, boosted by neural networks predicting the moves.  ...  Introduction The game of Go has been around for centuries and it is indeed a very popular brainteaser nowadays. Along with Chess and Checkers, Go is a game of perfect information.  ...  of the Go game tree.  ... 
dblp:conf/ysip/HolldoblerMT17 fatcat:vlidmhikyjdapn2zqobr2azgz4

Move Evaluation in Go Using Deep Convolutional Neural Networks [article]

Chris J. Maddison, Aja Huang, Ilya Sutskever, David Silver
2015 arXiv   pre-print
When the trained convolutional network was used directly to play games of Go, without any search, it beat the traditional search program GnuGo in 97% of games, and matched the performance of a state-of-the-art  ...  The game of Go is more challenging than other board games, due to the difficulty of constructing a position or move evaluation function.  ...  Combining MCTS with a large deep neural network is far from trivial, since the CNN is slower than the natural speed of the search, and it is not feasible to evaluate every node with the neural network.  ... 
arXiv:1412.6564v2 fatcat:twhvr332fzgtlk36ttij5cp2pi

Advances in Computer Go
コンピュータ囲碁の進歩

Hitoshi Matsubara
2017 Journal of the Robotics Society of Japan  
Graepel and D. Hassabis: "Mastering the Game of Go with Deep Neural Networks and Tree Search," Nature, vol. ] A.L.  ...  Zobrist: "A model of visual organization for the game of Go," AFIPS Spring Joint Computer Conference, vol.34, pp.103-112, 1969. [ 3 ] W. Reitman and B.  ... 
doi:10.7210/jrsj.35.191 fatcat:dgfajwnejndlfa5cu4ailao7rm

Exploring the Performance of Deep Residual Networks in Crazyhouse Chess [article]

Sun-Yu Gordon Chi
2019 arXiv   pre-print
We propose the novel creation of a neural network-based evaluation function for Crazyhouse.  ...  Until 2018, all competitive computer engines for this board game made use of an alpha-beta pruning algorithm with a hand-crafted evaluation function for each position.  ...  Then, using this a base checkpoint, we improve the network by reinforcing the network with self-play games generated with the help of the Monte Carlo Tree Search.  ... 
arXiv:1908.09296v1 fatcat:l765pd2sqncfnbz7tppqf4cjre

Mastering board games

Murray Campbell
2018 Science  
The approach described by Silver et al. augments deep reinforcement learning with a general-purpose searching method, Monte Carlo tree search (MCTS) (10) .  ...  More recently, deep (many-layer) neural networks were combined with reinforcement learning in an approach dubbed "deep reinforcement learning," which received widespread interest after it was successfully  ...  Klinger and G. Tesauro Contemplating the next move In the game between AlphaZero (white) and Stockfish (black), there were several moves that were reasonable for AlphaZero to consider.  ... 
doi:10.1126/science.aav1175 pmid:30523099 fatcat:woerdtqu6fgtldmtia2taq6i6y

The Game Is Not over Yet—Go in the Post-AlphaGo Era

Attila Egri-Nagy, Antti Törmänen
2020 Philosophies  
Here, we investigate these and related questions with respect to the special properties of Go (meaningful draws and extreme combinatorial complexity).  ...  The game of Go was the last great challenge for artificial intelligence in abstract board games.  ...  Neural networks without search can already be very strong, but combined with a Monte Carlo tree search they can reach superhuman level. Why does deep learning work?  ... 
doi:10.3390/philosophies5040037 fatcat:dsm72oixp5bdhcha7orgayrzuu

Multiplayer AlphaZero [article]

Nick Petosa, Tucker Balch
2019 arXiv   pre-print
Our experiments show that multiplayer AlphaZero learns successfully and consistently outperforms a competing approach: Monte Carlo tree search.  ...  Our work supports the use of AlphaZero in multiplayer games and suggests future research for more complex environments.  ...  Both v and p inform a Monte Carlo tree search (MCTS) to guide search over the game tree.  ... 
arXiv:1910.13012v3 fatcat:npceip42vzdffgihhf7qj5ispq

Artificial Intelligence Technology for Industrial Robot Applications
人工知能技術のロボット産業応用

Shohei Hido
2017 Journal of the Robotics Society of Japan  
Graepel and D. Hassabis: "Mastering the Game of Go with Deep Neural Networks and Tree Search," Nature, no.529, pp.484-489, 2016. [ 3 ] V. Mnih, K. Kavukcuoglu, D. Silver, A. Graves, I. Antonoglou, D.  ...  Wierstra and M. Riedmiller: "Playing atari with deep reinforcement learning," NIPS Deep Learning Workshop, 2013. [ 4 ] S. Gu, E. Holly, T. Lillicrap and S.  ... 
doi:10.7210/jrsj.35.186 fatcat:z7pjkqk6efe2poz6tkody2xbeu

Hyper-Parameter Sweep on AlphaZero General [article]

Hui Wang, Michael Emmerich, Mike Preuss, Aske Plaat
2019 arXiv   pre-print
Since AlphaGo and AlphaGo Zero have achieved breakground successes in the game of Go, the programs have been generalized to solve other tasks.  ...  We use the game of play 6×6 Othello, on the AlphaZeroGeneral open source re-implementation of AlphaZero.  ...  Hui Wang acknowledges financial support from the China Scholarship Council (CSC), CSC No.201706990015.  ... 
arXiv:1903.08129v1 fatcat:ostlc34i3nan3eajxj36nveqwa
« Previous Showing results 1 — 15 out of 4,154 results