hai-h-nguyen / rl-paper-collection Goto Github PK

Contain notes on paper on Reinforcement Learning that I read. Based on this list from Spinup OpenAI.

• Playing Atari with Deep Reinforcement Learning, Mnih et al, 2013. Algorithm: DQN. paper note
• Domain Adaptation for reinforcement learning on the atari paper

• Policy invariance under reward transformations: Theory and application to reward shaping. paper

• Data-Efficient Hierarchical Reinforcement Learning paper note

• ICML2015 - Universal Value Function Approximators

• Deep Reinforcement Learning that matters

• Between MDPs and semi-MDPs: A framework for temporal abstraction in reinforcement learning
• Pixel-Attentive Policy Gradient for Multi-Fingered Grasping in Cluttered Scences
• kPAM: KeyPoint Affordances for Category-Level Robotic Manipulation
• MAT: Multi-fingered adaptive tactile grasping via deep reinforcement learning
• Self-supervised correspondence in visuomotor policy learning

• Learning to Manipulate Object Collections Using Grounded State Representations (CoRL 2019) [note]
• Teacher algorithms for curriculum learning of Deep RL in continuously parameterized environments (CoRL 2019) [note]

• Playing FPS Games with DRL note
• Visual Reinforcement Learning with Imagined Goals note
• Deep Recurrent Q-Learning for POMDPs note

• Deep Q-learning from Demonstrations [note] [paper]
• Deep Reinforcement Learning with Double Q-Learning [note] paper
• Dueling Network Architectures for Deep Reinforcement Learning [note] [paper]
• Learning Latent Plans from Play [note] [paper] [project-site] CoRL2019
• Time Limit in RL [note] [paper] ICML 2018
• Multi-model imitation learning in partially observable environments [note] AAMAS 2020 (extended abstract)
• Learning belieft representations for imitation learning in pomdps [note] [code] Alg: Belief-module imitation learning (BMIL), UAI 2019
• Learning deep policies for robot bin picking by simulating robust grasping sequences [note] CORL 2017

• Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks a.k.a Cycle GAN [slide]