Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2511.12089 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912712004468736 |
|---|---|
| author | Fu, Yanchang Yin, Qiyue Liu, Shengda Xu, Pei Huang, Kaiqi |
| author_facet | Fu, Yanchang Yin, Qiyue Liu, Shengda Xu, Pei Huang, Kaiqi |
| contents | Excessive abstraction is a critical challenge in hand abstraction-a task specific to games like Texas hold'em-when solving large-scale imperfect-information games, as it impairs AI performance. This issue arises from extreme implementations of imperfect-recall abstraction, which entirely discard historical information. This paper presents KrwEmd, the first practical algorithm designed to address this problem. We first introduce the k-recall winrate feature, which not only qualitatively distinguishes signal observation infosets by leveraging both future and, crucially, historical game information, but also quantitatively captures their similarity. We then develop the KrwEmd algorithm, which clusters signal observation infosets using earth mover's distance to measure discrepancies between their features. Experimental results demonstrate that KrwEmd significantly improves AI gameplay performance compared to existing algorithms. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_12089 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | KrwEmd: Revising the Imperfect-Recall Abstraction from Forgetting Everything Fu, Yanchang Yin, Qiyue Liu, Shengda Xu, Pei Huang, Kaiqi Artificial Intelligence Computer Science and Game Theory Excessive abstraction is a critical challenge in hand abstraction-a task specific to games like Texas hold'em-when solving large-scale imperfect-information games, as it impairs AI performance. This issue arises from extreme implementations of imperfect-recall abstraction, which entirely discard historical information. This paper presents KrwEmd, the first practical algorithm designed to address this problem. We first introduce the k-recall winrate feature, which not only qualitatively distinguishes signal observation infosets by leveraging both future and, crucially, historical game information, but also quantitatively captures their similarity. We then develop the KrwEmd algorithm, which clusters signal observation infosets using earth mover's distance to measure discrepancies between their features. Experimental results demonstrate that KrwEmd significantly improves AI gameplay performance compared to existing algorithms. |
| title | KrwEmd: Revising the Imperfect-Recall Abstraction from Forgetting Everything |
| topic | Artificial Intelligence Computer Science and Game Theory |
| url | https://arxiv.org/abs/2511.12089 |