ChessLab
A chess learner that starts from random weights and is never told anything about chess.
The learner gets the rules of chess and nothing else: no openings, no piece values, no engine scores as targets. It plays itself, trains a candidate on its own games, and the candidate only becomes the new champion if it beats the old one over enough games to prove it.
Everything is measured against something the learner cannot author. Fixed opponents that never drift, frozen mate positions, and its own past generations, archived so they can be replayed.
The interesting findings are the ones that contradicted the plan. A network sixteen times bigger learned nothing more. Half the time went to Python’s garbage collector. And the search was the ceiling, not the model: the same network solves a third more mates when it is allowed to think longer.
- Opponent ladder grows to twenty-one rungs
- Bot opens to the public, gets a rating
- chesslab.dk goes live; bot plays Lichess
- Search 2.2x cheaper per node; mate probe climbs
- Learner spars with outside opponents, not itself
- Fixed reference ladder gives first absolute score
- Self-play learner with tree reuse and champions
- Four times the thinking, exactly the same progress per hour
The learner searched 48 positions per move before storing what it learned from. I raised it to 200 and let it run a day. Per cycle it…