DeepAI
Log In Sign Up

Stochastic Neural Networks for Hierarchical Reinforcement Learning

04/10/2017
by   Carlos Florensa, et al.
0

Deep reinforcement learning has achieved many impressive results in recent years. However, tasks with sparse rewards or long horizons continue to pose significant challenges. To tackle these important problems, we propose a general framework that first learns useful skills in a pre-training environment, and then leverages the acquired skills for learning faster in downstream tasks. Our approach brings together some of the strengths of intrinsic motivation and hierarchical methods: the learning of useful skill is guided by a single proxy reward, the design of which requires very minimal domain knowledge about the downstream tasks. Then a high-level policy is trained on top of these skills, providing a significant improvement of the exploration and allowing to tackle sparse rewards in the downstream tasks. To efficiently pre-train a large span of skills, we use Stochastic Neural Networks combined with an information-theoretic regularizer. Our experiments show that this combination is effective in learning a wide span of interpretable skills in a sample-efficient way, and can significantly boost the learning performance uniformly across a wide range of downstream tasks.

READ FULL TEXT

page 7

page 8

page 9

page 10

page 15

page 17

10/10/2019

Hierarchical Reinforcement Learning with Advantage-Based Auxiliary Rewards

Hierarchical Reinforcement Learning (HRL) is a promising approach to sol...
10/20/2021

Hierarchical Skills for Efficient Exploration

In reinforcement learning, pre-trained low-level skills have the potenti...
04/29/2022

Unsupervised Reinforcement Learning for Transferable Manipulation Skill Discovery

Current reinforcement learning (RL) in robotics often experiences diffic...
11/23/2022

Choreographer: Learning and Adapting Skills in Imagination

Unsupervised skill learning aims to learn a rich repertoire of behaviors...
09/30/2022

ASPiRe:Adaptive Skill Priors for Reinforcement Learning

We introduce ASPiRe (Adaptive Skill Prior for RL), a new approach that l...
12/01/2019

Affect-based Intrinsic Rewards for Learning General Representations

Positive affect has been linked to increased interest, curiosity and sat...
12/26/2022

Toward Efficient Automated Feature Engineering

Automated Feature Engineering (AFE) refers to automatically generate and...