{"schema":"postcutoff/event@1","as_of":"2026-11-22T13:32:00+01:00","url":"https://postcutoff.com/e/2026-09-30-ataraxos-stratego-nature/","md":"https://postcutoff.com/e/2026-09-30-ataraxos-stratego-nature/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"2026-09-30-ataraxos-stratego-nature","date":"2026-09-30","date_precision":"day","short_title":"Ataraxos beats Stratego's top player 15-1-4 at a fraction of DeepNash's compute","deck":null,"takeaway":"On Sept 30, 2026 Nature published Ataraxos, a Stratego AI from researchers at MIT, CMU, NYU and Stanford led by MIT's Gabriele Farina.","category":"research","category_label":"Research","importance":3,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":[{"n":1,"title":"Nature: Scalable decision-making for games of imperfect information","url":"https://www.nature.com/articles/s41586-026-11036-y","type":"paper","group":"primary","domain":"nature.com"},{"n":2,"title":"Ars Technica: AI finally beat the best Stratego player in history, and did it on a budget","url":"https://arstechnica.com/science/2026/10/ai-finally-beat-the-best-stratego-player-in-history-and-did-it-on-a-budget/","type":"press","group":"press","domain":"arstechnica.com"},{"n":3,"title":"SiliconSnark: MIT's Ataraxos beats Stratego champions","url":"https://www.siliconsnark.com/mits-ataraxos-beats-stratego-champions-without-requisitioning-a-small-power-plant/","type":"press","group":"press","domain":"siliconsnark.com"},{"n":4,"title":"AI Weekly: Ataraxos AI tops Stratego champion for a few thousand dollars","url":"https://aiweekly.co/alerts/ataraxos-ai-tops-stratego-champion-for-a-few-thousand-dollars","type":"press","group":"press","domain":"aiweekly.co"}],"official":1,"filed":"2026-10-01","updated":"2026-10-04","orgs":["MIT","Carnegie Mellon University","NYU","Stanford University"],"title":"Ataraxos beats Stratego's top player 15-1-4 at a fraction of DeepNash's compute (Nature)","summary":"On Sept 30, 2026 Nature published Ataraxos, a Stratego AI from researchers at MIT, CMU, NYU and Stanford led by MIT's Gabriele Farina. It beat Pim Niemeijer, the game's most decorated player, 15 wins, 1 loss and 4 draws, and went about 38–39 wins to 2 losses against elite players at a world championship exhibition. It combines self-play RL with decision-time planning over a learned belief model of hidden pieces. It beat DeepMind's DeepNash using under 1% of its training examples and a few thousand dollars of GPU time.","key_facts":["Paper: 'Scalable decision-making for games of imperfect information', Nature, Sept 30, 2026 (preprint on arXiv Nov 2025)","Lead: Gabriele Farina (MIT EECS); collaborators at Carnegie Mellon, NYU and Stanford","Vs Pim Niemeijer: 15 wins, 1 loss, 4 draws over 20 games","World-championship exhibition vs elite players: reported as 38-2 (MIT News via SiliconSnark) or 39-2 (other reports)","Reaches 'strictly higher playing strength than DeepNash while using less than one hundredth of the training examples'; the paper estimates DeepNash would cost $3–4.5M to train today","Training: 16 NVIDIA H100s for about a week, plus 4 H100s for 4 days for the belief network (reported)","Also superhuman Barrage Stratego, and state-of-the-art Hanabi and dou dizhu results"],"key_numbers":[],"tags":["games","imperfect-information","reinforcement-learning","search","stratego","nature"],"science":null,"body_md":"## What happened\n\nStratego hides each piece's identity, which makes classic game-tree search hard. DeepMind's DeepNash (2022) reached expert level\nwith huge compute and no search. Ataraxos instead plans at decision time over sampled hidden states from a generative belief network,\nand needs far less training.\n\n## Why it matters\n\nIt shows how cheap superhuman play in a large imperfect-information game has become. Explicit belief modelling plus search beats\nbrute-force self-play, a lesson relevant to agents that act under hidden information.\n\nUnverified: the exhibition record differs between reports (38-2 vs 39-2); the Nature paper itself was paywalled in this run.","disputed":[],"related":[],"people":[],"posts":[],"videos":[],"models":[],"changes":[{"date":"2026-10-04","type":"edit","text":"Added Ars Technica coverage (HN front page, 277 pts)"},{"date":"2026-10-01","type":"filed","text":"Created"}],"provenance":{"agents":[{"model":"Claude Opus 5.5","maker":"Anthropic","tool":"Claude Code"}],"filed":"2026-10-01","run":null,"sources_read":null,"updated":"2026-10-04","human_review":null,"version":{"date":"2026-10-04"}},"gaps":[{"model_id":"gpt-6-2-astra","name":"GPT-6.2 Astra","cutoff":"2026-05","days_after":122,"in_training_data":false},{"model_id":"claude-opus-5-5","name":"Claude Opus 5.5","cutoff":"2026-06","days_after":92,"in_training_data":false},{"model_id":"gemini-4-hydrogen","name":"Gemini 4 Hydrogen","cutoff":"2026-03","days_after":183,"in_training_data":false},{"model_id":"grok-4-7","name":"Grok 4.7","cutoff":"2026-05","days_after":122,"in_training_data":false}],"short_url":null}