Open-Source Projects

Aligner: Achieving Efficient Alignment through Weak-to-Strong Correction
a new efficient alignment paradigm, bypasses the whole RLHF process.
Last updated on Jun 17, 2023
PDF Code
Aligner: Achieving Efficient Alignment through Weak-to-Strong Correction
OmniSafe: An Infrastructure for Accelerating Safe Reinforcement Learning Research
a foundational framework designed to expedite SafeRL research endeavors
Last updated on Jun 17, 2023
PDF Code
OmniSafe: An Infrastructure for Accelerating Safe Reinforcement Learning Research
PKU-Beaver: Constrained Value-Aligned LLM via Safe RLHF
a highly modular open-source RLHF framework developed by the PKU-Alignment team
Last updated on Jun 17, 2023
PDF Code Video
PKU-Beaver: Constrained Value-Aligned LLM via Safe RLHF
Heterogeneous-Agent Reinforcement Learning
official implementation of Heterogeneous-Agent Reinforcement Learning (HARL) algorithms
Last updated on Jun 17, 2023
PDF Code
Heterogeneous-Agent Reinforcement Learning
MALib: A Parallel Framework for Population-based Reinforcement Learning
a parallel framework of population-based learning nested with reinforcement learning methods
Last updated on Jun 17, 2023
PDF Code
MALib: A Parallel Framework for Population-based Reinforcement Learning
MARLlib: A Scalable Multi-agent Reinforcement Learning Library
a comprehensive MARL algorithm library for solving multi-agent problems
Last updated on Jun 17, 2023
PDF Code
MARLlib: A Scalable Multi-agent Reinforcement Learning Library
TorchOpt: An Efficient PyTorch Library for Differentiable Optimization
a PyTorch-based efficient library for differentiable optimization
Last updated on Jun 17, 2023
PDF Code
TorchOpt: An Efficient PyTorch Library for Differentiable Optimization