A modular PyTorch library designed for learning, training, and deploying world models across various environments.
DreamerV3 World Model RL from Scratch — Educational implementation of model-based reinforcement learning
Curated papers, code, datasets, and benchmarks for medical world models in imaging, EHR trajectories, treatment planning, surgical AI, robotics, and virtual-cell simulation.
Component-level benchmark for catastrophic forgetting in world models. Two negative results: forgetting does not follow the labelled task-distance axis, and it happens in the encoder, where the usual metrics cannot see it. 375 runs, with code, data and paper.
Train a neural world model on your own video data, then play it live with WASD. Record → Kaggle → Play
双语技术博客 · 世界模型 / RSSM / DreamerV3 / VLA / Sim-to-Real
This project tests that hypothesis using two 1D physical phenomena — heat diffusion and Burgers equation — as elementary World Models, and advection-diffusion as the compound target phenomenon that neither model can represent alone.
A neural network learns Pac-Man, a world model learns to dream the game, and a third agent learns to play entirely inside those dreams. PPO → RSSM world model → imagination training → online dream loop. Custom vectorized NumPy engine, PyTorch, built from scratch.
Recurrent world models for time-series control, with multi-step rollout training to reduce compounding error.
End-to-end home robot navigation with DreamerV3 world models: RSSM latent imagination planning | 基于 DreamerV3 世界模型的居家机器人导航系统
Sonification and visualization of latent space deviations in industrial RSSM models — reconstructing tacit knowledge of skilled workers through data-driven perceptual transduction.
Official implementation for the paper Enhancing Eye Feature Estimation from Event Data Streams through Adaptive Inference State Space Modeling (ETRA '26).
RSSM-based world model for latent market dynamics, regime detection, and risk management. DreamerV2-style architecture.
RSSM world model on a POMDP pendulum, actor critic in imagination, and a representation ablation. Reconstruction measurably helps open loop prediction for 5 steps and not after.
World model (Dreamer/RSSM) vs. model-free PPO for cooperative multi-agent control on PettingZoo Multiwalker with shared policy
To associate your repository with the rssm topic, visit your repo's landing page and select "manage topics."