25,161
edits
Changes
no edit summary
'''Book Learning''',<br/>
is a technique of unsupervised [[Reinforcement Learning|reinforcement learning]] motivated by basic requirements for successfully playing a sequence of games, such as avoiding losing games in the same way from certain lines from the [[Opening Book|opening book]], or to recognize its own strong points and guide the engine into positions it does well <ref>[http://www.top-5000.nl/authors/rebel/hints.htm Learning Methods] by [[Ed Schroder|Ed Schröder]]</ref>.
Unsupervised reinforcement learning is characterized by [[Trial and Error|trial and error]] and feeding back reinforcement signals, in this context to modify move probabilities or to add new book moves. The signals are either based on the final outcome of the game, called result-driven learning, and/or by [[Search|search-driven]] learning, i.e. from the [[Score|score]] and trend of first few searches after leaving the book to determine how the program likes the position <ref>[[Robert Hyatt]] ('''1999'''). ''[http://www.craftychess.com/hyatt/learning.html Book Learning - a Methodology to Tune an Opening Book Automatically]''. [[ICGA Journal#22_1|ICCA Journal, Vol. 22, No. 1]]</ref>.
=External Links=
* [http[:Category:Stanley Clarke|Stanley Clarke]] - [https://wwwen.top-5000wikipedia.nlorg/wiki/authorsSchool_Days_(album) School Days] (1976), [https:/rebel/hintsen.wikipedia.htm Learning Methodsorg/wiki/YouTube YouTube] by [[Ed SchroderVideo: {{#evu:https://www.youtube.com/watch?v=Xk_gzDQd4zw|alignment=left|Ed Schröder]]valignment=top}}
=References=
'''[[Learning|Up one Level]]'''
[[Category:Stanley Clarke]]
