proroklab / rllib_differentiable_comms
This is a minimal example to demonstrate how multi-agent reinforcement learning with differentiable communication channels and centralized critics can be realized in RLLib. This example serves as a reference implementation and starting point for making RLLib more compatible with such architectures.
☆39Updated 11 months ago
Related projects: ⓘ
- Benchmarking RL generalization in an interpretable way.☆128Updated 7 months ago
- JAX and PZ RL envs + algorithms for swarms of CrazyFlies☆57Updated 3 weeks ago
- A tool for aggregating and plotting MARL experiment data.☆57Updated 3 weeks ago
- Partially Observable Process Gym☆158Updated 2 months ago
- ☆30Updated 3 years ago
- Source code for "A Policy Gradient Algorithm for Learning to Learn in Multiagent Reinforcement Learning" (ICML 2021)☆31Updated last year
- Implementation of Trajectory Transformer with attention caching and batched beam search☆101Updated last year
- Conservative Q Learning on top of SAC☆118Updated last year
- Datasets with baselines for offline multi-agent reinforcement learning.☆125Updated this week
- LAMBDA is a model-based reinforcement learning agent that uses Bayesian world models for safe policy optimization☆31Updated last year
- Simple maze environments using mujoco-py☆52Updated 8 months ago
- Author's PyTorch implementation of TD7 for online and offline RL☆108Updated last year
- Implementation of Truncated Quantile Critics method for continuous reinforcement learning. https://bayesgroup.github.io/tqc/☆90Updated 3 years ago
- Baseline implementation of recurrent PPO using truncated BPTT☆118Updated 4 months ago
- Official codebase for "B-Pref: Benchmarking Preference-BasedReinforcement Learning" contains scripts to reproduce experiments.☆108Updated 2 years ago
- JAX implementation of deep RL agents with resets from the paper "The Primacy Bias in Deep Reinforcement Learning"☆96Updated 2 years ago
- A pytorch reprelication of the model-based reinforcement learning algorithm MBPO☆150Updated 2 years ago
- Pytorch version of Dreamer, which follows the original TF v2 codes.☆112Updated 2 years ago
- ☆187Updated last year
- PyTorch implementation of the Option-Critic framework, Harb et al. 2016☆113Updated last month
- ☆192Updated 7 months ago
- 🔥 Datasets and env wrappers for offline safe reinforcement learning☆65Updated last week
- IMP-MARL: a Suite of Environments for Large-scale Infrastructure Management Planning via MARL☆35Updated last week
- Prioritized Experience Replay implementation with proportional prioritization☆67Updated last year
- Gridworld domains in the gym interface☆24Updated 10 months ago
- Evaluating long-term memory of reinforcement learning algorithms☆129Updated last year
- The Starcraft Multi-Agent challenge lite☆32Updated last week
- 🤖 Elegant implementations of offline safe RL algorithms in PyTorch☆161Updated last week
- Implementation of 'RL^2: Fast Reinforcement Learning via Slow Reinforcement Learning'☆54Updated 2 years ago
- An open-source framework to benchmark and assess safety specifications of Reinforcement Learning problems.☆61Updated last year