An empowerment-based solution to robotic manipulation tasks with sparse rewards
Abstract 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 whe...
Main Authors: | Dai, Siyu, Xu, Wei, Hofmann, Andreas, Williams, Brian |
---|---|
Other Authors: | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
Format: | Article |
Language: | English |
Published: |
Springer US
2023
|
Online Access: | https://hdl.handle.net/1721.1/151079 |
Similar Items
-
An Empowerment-based Solution to Robotic Manipulation Tasks with Sparse Rewards
by: Dai, Siyu, et al.
Published: (2022) -
Curriculum Learning for Robot Manipulation Tasks With Sparse Reward Through Environment Shifts
by: Erdi Sayar, et al.
Published: (2024-01-01) -
Fast-Reactive Probabilistic Motion Planning for High-Dimensional Robots
by: Dai, Siyu, et al.
Published: (2021) -
Learning to Make Decisions in Robotic Manipulation
by: Dai, Siyu
Published: (2022) -
Deep Deterministic Policy Gradient-Based Autonomous Driving for Mobile Robots in Sparse Reward Environments
by: Minjae Park, et al.
Published: (2022-12-01)