Michael Laskin

vqvae. A pytorch implementation of the vector quantized variational autoencoder (https://arxiv.org/abs/1711.00937)

curl. CURL: Contrastive Unsupervised Representation Learning for Sample-Efficient Reinforcement Learning

rad. RAD: Reinforcement Learning with Augmented Data

torchingup. TorchingUp provides minimal implementations of common Reinforcement Learning algorithms written in PyTorch. It is meant to complement OpenAI’s SpinningUp repository, which contains similar algorithms implemented in Tensorflow.

plan2explore. Python

vqvae2. bigger better vqvae

goal_sac. Python

gym-miniworld. Simple 3D interior simulator for RL & robotics research

ml-coursework. Assignments and notes for cs231: Convolutional Neural Networks and cs294: Deep Reinforcement Learning

rl-prep. Jupyter Notebook

vae. Variation autoencoder in pytorch

metaworld. An open source robotics benchmark for meta- and multi-task reinforcement learning

slate. Beautiful static documentation for your API

GANs. Implementing GANs in Keras

rlkit. Collection of reinforcement learning algorithms

spinningup. An educational resource to help anyone learn deep reinforcement learning.