DI-engine
OpenDILab Decision AI Engine. The Most Comprehensive Reinforcement Learning Framework B.P.
- 3.6k
- Python
- Apache-2.0
- 12 mo ago
OpenDILab Decision AI Engine. The Most Comprehensive Reinforcement Learning Framework B.P.
A modular, primitive-first, python-first PyTorch library for Reinforcement Learning.
A curated list of awesome model based RL resources (continually updated)
Implementation of Danijar's latest iteration for his Dreamer line of work
Deep active inference agents using Monte-Carlo methods
A curated list of awesome Model-based reinforcement learning resources
Soft Actor-Critic implementation with SOTA model-free extension (REDQ) and SOTA model-based extension (MBPO).
From next-token to next-state — a curated, opinionated map of world models, tagged by what each system predicts as the next state. Five-Axes framework, landscape map, and a live site.
Exploring the fundamentals of reinforcement learning (RL) to build agents capable of navigating complex real-world environments and enhancing the training of large language models (LLMs)
An "over-optimistic" effort to read and summarize a Deep Reinforcement Learning based paper a day 🤩 👊