Metraipolis › archive › dc_283fcf28da3a

Zero-Search Reinforcement Learning for Slay the Spire 2 with Failure-Driven Restarts — arXiv:2611.12088

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.12088
License (as archived): CC0-1.0
Archived: 2026-11-10T18:43:53.849Z
SHA-256 of the archived text: ad6aabea372305b0a49baf00454449ee68f9dcc6b6f6fa1afd0a2493e05e7479
Stable URL of this page: https://metraipolis.com/documento/dc_283fcf28da3a

How to cite

Zero-Search Reinforcement Learning for Slay the Spire 2 with Failure-Driven Restarts — arXiv:2611.12088 — arXiv (en), https://arxiv.org/abs/2611.12088 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_283fcf28da3a (Metraipolis archive dc_283fcf28da3a, sha256 ad6aabea372305b0a49baf00454449ee68f9dcc6b6f6fa1afd0a2493e05e7479)

BibTeX

@misc{dc283fcf28da3a,
  title = {Zero-Search Reinforcement Learning for Slay the Spire 2 with Failure-Driven Restarts — arXiv:2611.12088},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.12088},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_283fcf28da3a; sha256 ad6aabea372305b0a49baf00454449ee68f9dcc6b6f6fa1afd0a2493e05e7479},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.12088. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Slay the Spire 2 (STS2) is a deck-building roguelike in which a winning run at the highest difficulty takes several hundred decisions, and most losses happen in late boss fights that a learner seldom reaches. We train agents that choose every action with one forward pass of a 6.6M-parameter entity transformer, without lookahead. Training and evaluation run the game's own compiled assembly (v0.107.1) in a headless .NET host. The host replays a run exactly from its seed and actions, so training episodes can restart from states the learner has visited, and a failure-driven curriculum allocates these restarts in proportion to where complete runs are lost. On 12,800 held-out games at Ascension 10 (A10), the released policies win 8.05% averaged over the five characters (Ironclad 15.6%, Regent 12.5%), and AlphaSpire's released networks used without search win 0.55% on a common panel. Longer A10 training raised Ironclad from 12.4% to 15.6% and Defect from 3.0% to 8.1% on these games. At Ascension 0 the released policies win 80.2%, level with the 80.3% that the tree-search agent Spire Mentor reports, while Mentor leads by 5 to 8 points from A4 upward. Run in our host on the same 2,000 A10 games, Mentor's search wins 8.8% against 8.2% for our policies, which need 21-27 ms per decision on one CPU core. GPT-6 Astra at maximum reasoning won none of 11 A10 games through our text interface. In a controlled study of 18 training runs from six random initializations, failure-driven allocation exceeds uniform restart allocation on each of three Regent initializations, by 3.6 percentage points on average, while the Silent effect is small and changes sign. Code, trained policies, an in-game mod and the records behind every table are at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.12088 — 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 2026-11-10T18:37:37.676Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).