"RL Fine-Tuning: Scalable Online Planning via Reinforcement Learning Fine-Tuning", Fickinger et al 2021 {FB}

• KataGo

  49 3,235 9.3 C++

  GTP engine and self-play learning in Go

MCTS is hard to beat for chess/Go, but I'm increasingly convinced that MCTS is a heuristic that's overfit to perfect-info deterministic board games. Even within chess/Go, David Wu (creator of KataGo and now a researcher at FAIR) has pointed out to me several interesting failure cases for MCTS. I do think with further algorithmic improvements and hardware scaling, RL Fine-Tuning might overtake MCTS in chess/Go, but the real goal is to develop more general algorithms that can be used in a wide variety of environments.

• nnue-pytorch

  14 283 6.3 C++

  Stockfish NNUE (Chess evaluation) trainer in Pytorch

  

Getting SOTA in chess would be earth-shattering, especially since Stockfish has now adopted very light-weight NNs (called NNUE) and has doubled down on alpha-beta search, regaining the upper hand against A0 style programs.

• InfluxDB

  www.influxdata.com featured

  Power Real-Time Data Analytics at Scale. Get real-time insights from all types of time series data with InfluxDB. Ingest, query, and analyze billions of data points in real-time with unbounded cardinality.

  

Suggest a related project