Bootstrapping from Game Tree Search
Why is this work in the frame?
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
The three-model screen
all 1,000 screened works →All three models called this out of scope.
Machine learning algorithm for updating heuristic evaluation functions in game tree search.
It develops a machine-learning algorithm for chess evaluation.
AI algorithm for learning from game-tree search is computer science, not research studies.
Abstract
In this paper we introduce a new algorithm for updating the parameters of a heuristic evaluation function, by updating the heuristic towards the values computed by an alpha-beta search. Our algorithm differs from previous approaches to learning from search, such as Samuel's checkers player and the TD-Leaf algorithm, in two key ways. First, we update all nodes in the search tree, rather than a single node. Second, we use the outcome of a deep search, instead of the outcome of a subsequent search, as the training signal for the evaluation function. We implemented our algorithm in a chess program Meep, using a linear heuristic function. After initialising its weight vector to small random values, Meep was able to learn high quality weights from self-play alone. When tested online against human opponents, Meep played at a master level, the best performance of any chess program with a heuristic learned entirely from self-play.
Stored with the screening record, where it is evidence for the labels above.
The record
- Venue
- Topic
- Artificial Intelligence in Games
- Field
- Computer Science
- Canadian institutions
- University of Alberta
- Funders
- —
- Keywords
- Incremental heuristic searchEvaluation functionBeam searchHeuristicComputer scienceSearch algorithmArtificial intelligenceNull-move heuristicBootstrapping (finance)Search treeFunction (biology)Best-first searchOutcome (game theory)Tree (set theory)Node (physics)Machine learningAlgorithmMathematicsEngineering
- Has abstract in OpenAlex
- yes