Evolving a Roving Eye for Go Export

@incollection{citeulike:2678591, abstract = {Go remains a challenge for artificial intelligence. Currently, most machine learning methods tackle Go by playing on a specific fixed board size, usually smaller than the standard 19×19 board of the complete game. Because such techniques are designed to process only one board size, the knowledge gained through experience cannot be applied on larger boards. In this paper, a roving eye neural network is evolved to solve this problem. The network has a small input field that can scan boards of any size. Experiments demonstrate that (1) The same roving eye architecture can play on different board sizes, and (2) experience gained by playing on a small board provides an advantage for further learning on a larger board. These results suggest a potentially powerful new methodology for computer Go: It may be possible to scale up by learning on incrementally larger boards, each time building on knowledge acquired on the prior board.}, author = {Stanley, Kenneth O. and Miikkulainen, Risto}, citeulike-article-id = {2678591}, citeulike-linkout-0 = {http://www.springerlink.com/content/96y7lyycbj8k67ey}, citeulike-linkout-1 = {http://nn.cs.utexas.edu/downloads/papers/stanley.gecco04.pdf}, journal = {Genetic and Evolutionary Computation – GECCO 2004}, keywords = {evolution, ga, games}, pages = {1226--1238}, posted-at = {2008-04-16 18:38:46}, priority = {2}, title = {Evolving a Roving Eye for Go}, url = {http://www.springerlink.com/content/96y7lyycbj8k67ey}, year = {2004} }

TY - CHAP ID - citeulike:2678591 L3 - citeulike-article-id:2678591 N2 - Go remains a challenge for artificial intelligence. Currently, most machine learning methods tackle Go by playing on a specific fixed board size, usually smaller than the standard 19×19 board of the complete game. Because such techniques are designed to process only one board size, the knowledge gained through experience cannot be applied on larger boards. In this paper, a roving eye neural network is evolved to solve this problem. The network has a small input field that can scan boards of any size. Experiments demonstrate that (1) The same roving eye architecture can play on different board sizes, and (2) experience gained by playing on a small board provides an advantage for further learning on a larger board. These results suggest a potentially powerful new methodology for computer Go: It may be possible to scale up by learning on incrementally larger boards, each time building on knowledge acquired on the prior board. JF - Genetic and Evolutionary Computation – GECCO 2004 EP - 1238 TI - Evolving a Roving Eye for Go SP - 1226 KW - evolution KW - ga KW - games AU - Stanley, Kenneth AU - Miikkulainen, Risto PY - 2004/// UR - http://www.springerlink.com/content/96y7lyycbj8k67ey ER -