Metraipolis › archive › dc_c8ff14ea8885

Can AI Agents Learn Their Way to the Top? Evaluating Heuristic Learning in a Long-Running Game Agent Competition — arXiv:2610.12341

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.12341
License (as archived): CC0-1.0
Archived: 2026-10-09T05:50:39.005Z
SHA-256 of the archived text: fc2be662fa0582a229f6ee3e15b60c109045709f623557d9651614d4432059ad
Stable URL of this page: https://metraipolis.com/documento/dc_c8ff14ea8885

How to cite

Can AI Agents Learn Their Way to the Top? Evaluating Heuristic Learning in a Long-Running Game Agent Competition — arXiv:2610.12341 — arXiv (en), https://arxiv.org/abs/2610.12341 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c8ff14ea8885 (Metraipolis archive dc_c8ff14ea8885, sha256 fc2be662fa0582a229f6ee3e15b60c109045709f623557d9651614d4432059ad)

BibTeX

@misc{dcc8ff14ea8885,
  title = {Can AI Agents Learn Their Way to the Top? Evaluating Heuristic Learning in a Long-Running Game Agent Competition — arXiv:2610.12341},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.12341},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c8ff14ea8885; sha256 fc2be662fa0582a229f6ee3e15b60c109045709f623557d9651614d4432059ad},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.12341. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Adversarial games have driven advances from heuristic search to reinforcement learning, yet learning and adapting strategies from limited samples remain challenging. AI agents offer an alternative by turning game experience into revisions of executable policies. Building on heuristic learning (HL), we formalize Adversarial Heuristic Learning (AHL), a paradigm that uses AI agents as learning engines to refine game policies and supporting software while keeping model weights fixed. We introduce AAArena, a benchmark comprising 12 authentic adversarial games and 1,920 archived human programs, with an evaluation protocol modeled on real-world game competitions. Agents interpret rules, choose opponents, analyze replays, and revise game agents to achieve their highest ranking within fixed match and evaluation budgets. We evaluate \val{completedmodels} model and harness configurations: Opus5.5 with Claude Code earns 6 gold medals, while no evaluated configuration tops the remaining 6 human ladders. Performance is generally weaker in games with more complex rule specifications. Further experiments show that opponent selection and dense feedback support policy improvement, and that agents learn from both on-policy replays of their own matches and off-policy replays of other players' matches. These results highlight HL's potential in adversarial games and identify persistent challenges in game understanding, strategy implementation, and long-horizon policy development.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.12341 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-09T05:47:01.106Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).