PyTorch reimplementation of "Deep Hierarchical Planning" RL framework. Features a multi-model architecture with manager-worker policies, world model, and goal autoencoder. Built with Python/PyTorch and includes experiment logging with Weights & Biases.
pythonmachine-learningreinforcement-learningpytorchdeeplearningresearch-paperworld-modelsrl-frameworkneurips-2022hierarchical-rlmanager-workergoal-autoencoder
-
Updated
Sep 10, 2025 - Python