opendilab / LightZero

[NeurIPS 2023 Spotlight] LightZero: A Unified Benchmark for Monte Carlo Tree Search in General Sequential Decision Scenarios (awesome MCTS)
https://huggingface.co/spaces/OpenDILabCommunity/ZeroPal
Apache License 2.0
1.15k stars 120 forks source link

feature(wrh): add soft modulization in unizero #246

Closed ruiheng123 closed 3 months ago

ruiheng123 commented 4 months ago

Soft Modulization with a route network allocating probablistic weights between one module in the previous layer and another module in the next layer. Paper link: https://arxiv.org/abs/2003.13661

ruiheng123 commented 4 months ago

This is a new version of implementing soft modulization (debugging).