Multi-Goal Reinforcement Learning: Challenging Robotics Environments and Request for Research (2018-02-26T00:00:00.000000Z)

TL;DR

A suite of challenging continuous control tasks (integrated with OpenAI Gym) based on currently existing robotics hardware and following a Multi-Goal Reinforcement Learning (RL) framework are introduced.

Abstract

The purpose of this technical report is two-fold. First of all, it introduces a suite of challenging continuous control tasks (integrated with OpenAI Gym) based on currently existing robotics hardware. The tasks include pushing, sliding and pick & place with a Fetch robotic arm as well as in-hand object manipulation with a Shadow Dexterous Hand. All tasks have sparse binary rewards and follow a Multi-Goal Reinforcement Learning (RL) framework in which an agent is told what to do using an additional input. The second part of the paper presents a set of concrete research ideas for improving RL algorithms, most of which are related to Multi-Goal RL and Hindsight Experience Replay.

Authors

Jonas Schneider

4 papers

Vikash Kumar

7 papers

Wojciech Zaremba

9 papers

Multi-Goal Reinforcement Learning: Challenging Robotics Environments and Request for Research

TL;DR

Abstract

Authors

References21 items

Temporal Difference Models: Model-Free Deep RL for Model-Based Control

Hierarchical Actor-Critic

Hindsight policy gradients

Proximal Policy Optimization Algorithms

Reverse Curriculum Generation for Reinforcement Learning

A Distributional Perspective on Reinforcement Learning

Hindsight Experience Replay

Parameter Space Noise for Exploration

Interpolated Policy Gradient: Merging On-Policy and Off-Policy Gradient Estimation for Deep Reinforcement Learning

Equivalence Between Policy Gradients and Soft Q-Learning

Learning to Play in a Day: Faster Deep Reinforcement Learning by Optimality Tightening

Safe and Efficient Off-Policy Reinforcement Learning

OpenAI Gym

Prioritized Experience Replay

Continuous control with deep reinforcement learning

Universal Value Function Approximators

Human-level control through deep reinforcement learning

Adam: A Method for Stochastic Optimization

MuJoCo: A physics engine for model-based control

OpenAI Baselines

Notice that normally off-policy data used by RL algorithms come from an earlier version of the policy and therefores is relatively close to on-policy data

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names