Offline RL on 5TB of Human Play Learns to Recommend Better Chess Puzzles
allenainie · x · 2026-08-18
The authors trained an offline RL policy on nearly 5TB of human behavior data (a full year of play) to recommend chess puzzles with high pedagogical value. After many challenges, they found a recipe that worked, and blind evaluation by International Masters and Grandmasters showed the policy recommends puzzles players find more fun and more challenging.
Related event: Offline RL Mines 5TB of Chess Games to Find High-Value Puzzles(2 posts)→
More from Research
- Self-organized Boolean Computation via Neural Cellular Automata — zzznah · 2026-08-18
- Ultra-fast Fourier transform and optical AI with a single lens — MeasurementDull7350 · 2026-08-18
- Developer releases playground to visualize how LLM text watermarking works — grabcard · 2026-08-18
- Giving the Bristlebot a Brain: First Agentic Micro-Robot That Senses, Predicts, and Acts — maier_ak · 2026-08-18
- Defining Transformer: architecture mixing inputs across feature and sequence dimensions — PandaAshwinee · 2026-08-18
- Building distributed training framework from scratch — hkproj · 2026-08-18