AI finally masters Stratego—on a budget

Kari Jaaskelainen

11 Nov 2025 — 1 min read

AI finally masters Stratego—on a budget

Stratego has long been a worst-case test for AI: you must plan deeply while most pieces are hidden. Big-budget attempts still fell short of top humans.

This paper reports a step change: an AI that reaches vastly superhuman Stratego performance—trained for only a few thousand dollars.

Self-play reinforcement learning: the system learns by playing itself, discovering tactics and long-term plans without human data.
Test-time search with hidden info: a lookahead that reasons about uncertainty, not just perfect knowledge.
General methods: designed for imperfect-information problems beyond games.

Why it matters: strong decision-making under uncertainty, at low cost, opens doors for research and applications in security, logistics, and robotics.

Authors: Samuel Sokota, Eugene Vinitsky, Hengyuan Hu, J. Zico Kolter, Gabriele Farina. Paper: http://arxiv.org/abs/2511.07312v1

Paper: http://arxiv.org/abs/2511.07312v1

Register: https://www.AiFeta.com

AI ReinforcementLearning Stratego GameAI MachineLearning ImperfectInformation Research SelfPlay Arxiv

Automating GDPR Compliance: A Roadmap for Companies and Law Firms

GDPR compliance is more than checkboxes. A new roadmap from the Privatech project shows how automation and machine learning can help companies and law firms assess—and even generate—privacy compliance. * Shift the focus to data processors’ real workflows: drafting policies, mapping data uses, documenting decisions. * Break compliance into machine-ready

FPGAs for Faster, Leaner Deep Learning: A Review of CNN Accelerators

Deep learning drives image search, robots, and medical scans. Most systems lean on CPUs and GPUs. This review asks: what if we run convolutional neural networks (CNNs) on FPGAs—reconfigurable chips you can tailor to the model? * Why FPGAs: custom dataflows, low latency, and strong energy efficiency—great for cameras,

Dynamic-K: Recommendations That Know When to Stop

Most apps show a fixed number of “top” items—say 10 movies or 20 products—assuming there are always enough good options. But that’s not always true: sometimes there are few relevant items, or some users are extra picky. The result? Filler recommendations. Dynamic-K flips the script. Instead of

Teaching chatbots to stop contradicting themselves (DECODE)

Teaching chatbots to stop contradicting themselves Ever had a bot say one thing, then the opposite a few turns later? This study introduces DECODE—a new task and dataset for spotting contradictions in everyday conversations, drawn from both human-human and human-bot chats. * New data beats existing natural language inference (NLI)

AI finally masters Stratego—on a budget

Read more

Automating GDPR Compliance: A Roadmap for Companies and Law Firms

FPGAs for Faster, Leaner Deep Learning: A Review of CNN Accelerators

Dynamic-K: Recommendations That Know When to Stop

Teaching chatbots to stop contradicting themselves (DECODE)