DeepAI
Log In Sign Up

An Empowerment-based Solution to Robotic Manipulation Tasks with Sparse Rewards

10/15/2020
by   Siyu Dai, et al.
0

In order to provide adaptive and user-friendly solutions to robotic manipulation, it is important that the agent can learn to accomplish tasks even if they are only provided with very sparse instruction signals. To address the issues reinforcement learning algorithms face when task rewards are sparse, this paper proposes a novel form of intrinsic motivation that can allow robotic manipulators to learn useful manipulation skills with only sparse extrinsic rewards. Through integrating and balancing empowerment and curiosity, this approach shows superior performance compared to other existing intrinsic exploration approaches during extensive empirical testing. Qualitative analysis also shows that when combined with diversity-driven intrinsic motivations, this approach can help manipulators learn a set of diverse skills which could potentially be applied to other more complicated manipulation tasks and accelerate their learning process.

READ FULL TEXT

page 6

page 8

06/23/2020

ELSIM: End-to-end learning of reusable skills through intrinsic motivation

Taking inspiration from developmental learning, we present a novel reinf...
09/30/2019

Efficient Bimanual Manipulation Using Learned Task Schemas

We address the problem of effectively composing skills to solve sparse-r...
10/07/2022

Generative Augmented Flow Networks

The Generative Flow Network is a probabilistic framework where an agent ...
05/31/2021

Q-attention: Enabling Efficient Learning for Vision-based Robotic Manipulation

Despite the success of reinforcement learning methods, they have yet to ...
11/30/2018

Modulated Policy Hierarchies

Solving tasks with sparse rewards is a main challenge in reinforcement l...
11/13/2020

Robotic self-representation improves manipulation skills and transfer learning

Cognitive science suggests that the self-representation is critical for ...