Self Supervised Reinforcement Learning

Self-supervised reinforcement learning (SSRL) aims to train reinforcement learning agents without relying on extensive labeled data, leveraging the inherent structure of the environment or task for learning. Current research focuses on developing effective self-supervised pre-training methods, often employing techniques like contrastive learning, clustering, or synthetic data generation to improve sample efficiency and generalization across tasks. These advancements are impacting various fields, including robotics, computer vision (e.g., image deraining), and knowledge graph reasoning, by enabling the development of more robust and data-efficient algorithms for complex decision-making problems. The ability to learn effectively from unlabeled data significantly reduces the need for human annotation, making SSRL a crucial step towards more scalable and practical AI systems.

Papers

May 22, 2024

Knowledge Graph Reasoning with Self-supervised Reinforcement Learning
Ying Ma, Owen Burns, Mingqiu Wang, Gang Li, Nan Du, Laurent El Shafey, Liqiang Wang, Izhak Shafran, Hagen Soltau
Reinforcement Learning Knowledge Graph Reasoning Incomplete Knowledge Graph Self Supervised Reinforcement Learning

May 20, 2024

Learning Future Representation with Synthetic Observations for Sample-efficient Reinforcement Learning
Xin Liu, Yaran Chen, Dongbin Zhao
Synthetic Data Visual Reinforcement Learning Sample Efficient Reinforcement Learning Downstream Policy Learning Self Supervised Reinforcement Learning Auxiliary Function Future Latent

March 28, 2024

Removing the need for ground truth UWB data collection: self-supervised ranging error correction using deep reinforcement learning
Dieter Coppens, Ben Van Herbruggen, Adnan Shahid, Eli De Poorter
Deep Reinforcement Learning Community Need Ultra Wideband Error Correction Indoor Positioning Self Supervised Reinforcement Learning

March 27, 2024

Image Deraining via Self-supervised Reinforcement Learning
He-Hao Liao, Yan-Tsung Peng, Wen-Tao Chu, Ping-Chun Hsieh, Chung-Chi Tsai
Image Deraining Self Supervised Reinforcement Learning

February 23, 2024

Foundation Policies with Hilbert Representations
Seohong Park, Tobias Kreiman, Sergey Levine
Reinforcement Learning Flawed Foundation Goal Conditioned Reinforcement Learning Self Supervised Objective Unsupervised Version Self Supervised Reinforcement Learning 2 Dimensional Hilbert

November 7, 2023

SeRO: Self-Supervised Reinforcement Learning for Recovery from Out-of-Distribution Situations
Chan Kim, Jaekyung Cho, Christophe Bobda, Seung-Woo Seo, Seong-Woo Kim
Reinforcement Learning Prompt Recovery Robotic Agent Understanding Out of Distribution Class Out of Distribution Learning Probability Distribution Self Supervised Reinforcement Learning

October 9, 2023

Integrating Stock Features and Global Information via Large Language Models for Enhanced Stock Return Prediction
Yujie Ding, Shuai Jia, Tianyi Ma, Bingcheng Mao, Xiuze Zhou, Liuliu Li, Dongming Han
Stock Return Prediction Global Information Self Supervised Reinforcement Learning Quantitative Investment

June 6, 2023

Stabilizing Contrastive RL: Techniques for Robotic Goal Reaching from Offline Data
Chongyi Zheng, Benjamin Eysenbach, Homer Walke, Patrick Yin, Kuan Fang, Ruslan Salakhutdinov, Sergey Levine
Reinforcement Learning Self Supervised Learning Self Supervised Barzilai Borwein Technique Offline Data Self Supervised Reinforcement Learning Robot Goal Contrastive Reinforcement Learning

May 26, 2023

Self-Supervised Reinforcement Learning that Transfers using Random Features
Boyuan Chen, Chuning Zhu, Pulkit Agrawal, Kaiqing Zhang, Abhishek Gupta
Formality Transfer Model Free Reinforcement Learning Random Feature Task Agnostic Self Supervised Reinforcement Learning

February 27, 2023

The Provable Benefits of Unsupervised Data Sharing for Offline Reinforcement Learning
Hao Hu, Yiqin Yang, Qianchuan Zhao, Chongjie Zhang
Offline Reinforcement Learning Offline Reinforcement Learning Algorithm Provable Benefit Reward Free Self Supervised Reinforcement Learning

February 12, 2023

Policy-Induced Self-Supervision Improves Representation Finetuning in Visual RL
Sébastien M. R. Arnold, Fei Sha
Individual Representation Representational Similarity Self Supervised Objective Self Supervised Reinforcement Learning Supervised Counterpart Encode Task

January 27, 2023

In-Distribution Barrier Functions: Self-Supervised Policy Filters that Avoid Out-of-Distribution States
Fernando Castañeda, Haruki Nishimura, Rowan McAllister, Koushil Sreenath, Adrien Gaidon
Learning Based Control Barrier Function State Space Visuo Motor Control Distribution State Self Supervised Reinforcement Learning

Self Supervised Reinforcement Learning

Papers

Knowledge Graph Reasoning with Self-supervised Reinforcement Learning

Learning Future Representation with Synthetic Observations for Sample-efficient Reinforcement Learning

Removing the need for ground truth UWB data collection: self-supervised ranging error correction using deep reinforcement learning

Image Deraining via Self-supervised Reinforcement Learning

Foundation Policies with Hilbert Representations

SeRO: Self-Supervised Reinforcement Learning for Recovery from Out-of-Distribution Situations

Integrating Stock Features and Global Information via Large Language Models for Enhanced Stock Return Prediction

Stabilizing Contrastive RL: Techniques for Robotic Goal Reaching from Offline Data

Self-Supervised Reinforcement Learning that Transfers using Random Features

The Provable Benefits of Unsupervised Data Sharing for Offline Reinforcement Learning

Policy-Induced Self-Supervision Improves Representation Finetuning in Visual RL

In-Distribution Barrier Functions: Self-Supervised Policy Filters that Avoid Out-of-Distribution States