awesome-deep-rl
github.com/tigerneil/awesome-deep-rl ↗For deep RL and the future of AI.
1.5k
GitHub Stars
73
Curated Resources
21
Categories
1 hour ago
Last Refreshed
General guidances2022General benchmark frameworksUnsupervisedOfflineValue basedActor-CriticModel-basedMulti-agentNew designMultitaskNo reward RLSeptember 2026 update: embodied RLEarlier foundations to retain (2023–2025)Canonical RL foundationsLLM and agentic RLEmbodied AI and real-world RLFoundation and world modelsOffline, hybrid, and data-centric RLMulti-agent coordinationGeneralization, safety, and evaluation
Use this list with your AI agent
Add the Context Awesome MCP server to Claude, Cursor, or any MCP client, then ask:
"Show me evaluation and adjacent embodied developments resources from awesome-deep-rl"
Installation instructions →What's inside
General benchmark frameworks
Unsupervised
- APS: Active Pretraining with Successor Feature
- Behavior From the Void: Unsupervised Active Pre-Training
- Diversity is All You Need: Learning Skills without a Reward Function
- Efficient Exploration via State Marginal Matching
- Exploration by Random Network Distillation
- Reinforcement Learning with Prototypical Representations
2022
General guidances
Value based
Showing a sample of 73 resources. View the full list on GitHub →