用对比学习训练棋盘嵌入空间，让模型像人一样直觉选步。
Learning to Plan via Supervised Contrastive Learning and Strategic Interpolation: A Chess Case Study
- 通过对比学习构建棋局的语义嵌入空间，距离表征评估相似性。
- 仅用6步广度搜索即达2593等级分，接近顶尖引擎表现。
- 适合研究类人决策、可迁移至其他完美信息博弈场景。
现代国际象棋引擎依靠深度树搜索与回归评估实现超人类水平，而人类棋手则依赖直觉选择候选着法，再通过浅层搜索验证。为模拟这种直觉式规划过程，我们使用监督对比学习训练一个Transformer编码器，将棋局状态嵌入到由位置评估结构化的潜在空间中。该空间中距离反映评估相似性，可视化轨迹显示了游戏状态间的可解释过渡。我们证明，移动选择可完全在嵌入空间内完成，通过向有利区域推进，无需依赖深度搜索。尽管仅采用6步束搜索，模型估计达到2593等级分。性能随模型规模和嵌入维度提升，表明潜在空间规划或可作为传统搜索的可行替代方案。虽然聚焦于国际象棋，但所提嵌入式规划方法可推广至其他具有可学习状态评估的完美信息博弈。所有源代码见 https://github.com/andrewhamara/SOLIS。
原文摘要 · Abstract (English)
Modern chess engines achieve superhuman performance through deep tree search and regressive evaluation, while human players rely on intuition to select candidate moves followed by a shallow search to validate them. To model this intuition-driven planning process, we train a transformer encoder using supervised contrastive learning to embed board states into a latent space structured by positional evaluation. In this space, distance reflects evaluative similarity, and visualized trajectories display interpretable transitions between game states. We demonstrate that move selection can occur entirely within this embedding space by advancing toward favorable regions, without relying on deep search. Despite using only a 6-ply beam search, our model achieves an estimated Elo rating of 2593. Performance improves with both model size and embedding dimensionality, suggesting that latent planning may offer a viable alternative to traditional search. Although we focus on chess, the proposed embedding-based planning method can be generalized to other perfect-information games where state evaluations are learnable. All source code is available at https://github.com/andrewhamara/SOLIS.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。