VRL3: A Data-Driven Framework for Visual Deep Reinforcement Learning

02/17/2022
by   Che Wang, et al.
0

We propose a simple but powerful data-driven framework for solving highly challenging visual deep reinforcement learning (DRL) tasks. We analyze a number of major obstacles in taking a data-driven approach, and present a suite of design principles, training strategies, and critical insights about data-driven visual DRL. Our framework has three stages: in stage 1, we leverage non-RL datasets (e.g. ImageNet) to learn task-agnostic visual representations; in stage 2, we use offline RL data (e.g. a limited number of expert demonstrations) to convert the task-agnostic representations into more powerful task-specific representations; in stage 3, we fine-tune the agent with online RL. On a set of highly challenging hand manipulation tasks with sparse reward and realistic visual inputs, our framework learns 370 previous SOTA method while using an encoder that is 50 times smaller, fully demonstrating the potential of data-driven deep reinforcement learning.

READ FULL TEXT
research
10/15/2020

Knowledge Transfer in Multi-Task Deep Reinforcement Learning for Continuous Control

While Deep Reinforcement Learning (DRL) has emerged as a promising appro...
research
12/01/2022

Launchpad: Learning to Schedule Using Offline and Online RL Methods

Deep reinforcement learning algorithms have succeeded in several challen...
research
09/28/2017

Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrations

Dexterous multi-fingered hands are extremely versatile and provide a gen...
research
11/15/2021

Piano Fingering with Reinforcement Learning

Hand and finger movements are a mainstay of piano technique. Automatic F...
research
05/26/2023

Physical Deep Reinforcement Learning: Safety and Unknown Unknowns

In this paper, we propose the Phy-DRL: a physics-model-regulated deep re...
research
04/24/2023

Efficient Halftoning via Deep Reinforcement Learning

Halftoning aims to reproduce a continuous-tone image with pixels whose i...
research
09/26/2019

A Framework for Data-Driven Robotics

We present a framework for data-driven robotics that makes use of a larg...

Please sign up or login with your details

Forgot password? Click here to reset