CST-RL: Contrastive Spatio-Temporal Representations for Reinforcement Learning

CST-RL: Contrastive Spatio-Temporal Representations for Reinforcement Learning

Learning representations from high-dimensional observations is critical for training of pixel-based continuous control tasks with reinforcement learning (RL). Without proper representations, the training will be very inefficient, requiring long training time and huge training data to learn directly...

Full description

Bibliographic Details
Main Authors: Chi-Kai Ho, Chung-Ta King
Format: Article
Language:English
Published: IEEE 2023-01-01
Series:IEEE Access
Subjects:
Reinforcement learning
3D CNNs
contrastive learning
spatio-temporal representation learning
sample efficiency
Online Access:https://ieeexplore.ieee.org/document/10075624/

Internet

https://ieeexplore.ieee.org/document/10075624/

Similar Items