Deepak Pathak

noreward-rl. [ICML 2017] TensorFlow code for Curiosity-driven Exploration for Deep Reinforcement Learning

context-encoder. [CVPR 2016] Unsupervised Feature Learning by Image Inpainting using GANs

pyflow. Fast, accurate and easy to run dense optical flow with python wrapper

unsupervised-video. [CVPR 2017] Unsupervised deep learning using unlabelled videos on the web

zeroshot-imitation. [ICLR 2018] TensorFlow code for zero-shot visual imitation by self-supervised exploration

videoseg. [CVPR 2017] Video motion segmentation and tracking

exploration-by-disagreement. [ICML 2019] TensorFlow Code for Self-Supervised Exploration via Disagreement

modular-assemblies. [NeurIPS 2019] Code for the paper "Learning to Control Self-Assembling Morphologies: A Study of Generalization via Modularity"

seg-by-interaction. Unsupervised instance segmentation via active robot interaction

ccnn. [ICCV 2015] Framework for optimizing CNNs with linear constraints for Semantic Segmentation

hierarchical-imitation. [NeurIPS 2019] Code for Hierarchical Third-Person Imitation Learning

caffe-ccnn. Branch of caffe for weakly supervised learning.

large-scale-curiosity-1. Code for the paper "Large-Scale Study of Curiosity-Driven Learning"

corr_clustering. Correlation Clustering Project

pathak22.github.io. HTML

dreamer. Dream to Control: Learning Behaviors by Latent Imagination

gym-pull. Add-on for OpenAI Gym that supports automatic downloading of user environments.

baselines. OpenAI Baselines: high-quality implementations of reinforcement learning algorithms