MAPLE algorithm enhances AlphaZero for imperfect-information games

Researchers have developed a new tree search method called MAPLE, designed to improve the performance of AlphaZero-style algorithms in imperfect-information games. Unlike previous methods that struggle with strategy fusion or high computational costs, MAPLE aggregates policy and value evaluations from multiple sampled world states within a single search tree. Experiments on Phantom Go and Dark Hex demonstrated significant Elo improvements, showing MAPLE's effectiveness for AlphaZero-like learning in complex games. AI

影响 Introduces a novel approach for AI agents to master complex games with incomplete information.

排序理由 Academic paper detailing a new algorithm for game AI. [lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

• AlphaZero

• 论文
• 其他

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →