Summary of Neural Network-based Information Set Weighting For Playing Reconnaissance Blind Chess, by Timo Bertram et al.
Neural Network-based Information Set Weighting for Playing Reconnaissance Blind Chess
by Timo Bertram, Johannes Fürnkranz, Martin Müller
First submitted to arxiv on: 8 Jul 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a novel approach to improving gameplay in imperfect information games, specifically in the context of Reconnaissance Blind Chess (RBC). By assigning weights to the states within an information set, the algorithm can reason about the likelihood of each state occurring during gameplay. Two neural networks are trained using historical game data to estimate these weights: a Siamese network and a classical convolutional network. The study finds that the Siamese network achieves higher accuracy and is more efficient than the convolutional network for this domain. An RBC-playing agent based on the generated weightings is evaluated, with parameter settings influencing how strongly it relies on them. The resulting best player ranks 5th on the public leaderboard. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper helps improve gameplay in a special kind of game called Reconnaissance Blind Chess. In this game, players don’t have all the information they need to make good moves. To deal with this, the algorithm tries to figure out which possible game states are most likely to happen. It uses two special kinds of computer models, Siamese and convolutional networks, to do this. The study finds that one model is better than the other for this particular game. An agent based on this approach plays RBC well enough to rank 5th on a leaderboard. |
Keywords
» Artificial intelligence » Convolutional network » Likelihood » Siamese network
