홈 › Papers

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

2023-06-06 · Chongyi Zheng, Benjamin Eysenbach, Homer Walke, Patrick Yin, Kuan Fang, Ruslan Salakhutdinov, Sergey Levine

Robotic systems that rely primarily on self-supervised learning have the potential to decrease the amount of human annotation and engineering effort required to learn control strategies. In the same way that prior robotic systems have leveraged self-supervised techniques from computer vision (CV) and natural language processing (NLP), our work builds on prior work showing that the reinforcement learning (RL) itself can be cast as a self-supervised problem: learning to reach any goal without human-specified rewards or labels. Despite the seeming appeal, little (if any) prior work has demonstrated how self-supervised RL methods can be practically deployed on robotic systems. By first studying a challenging simulated version of this task, we discover design decisions about architectures and hyperparameters that increase the success rate by $2 \times$. These findings lay the groundwork for our main result: we demonstrate that a self-supervised RL algorithm based on contrastive learning can solve real-world, image-based robotic manipulation tasks, with tasks being specified by a single goal image provided after training.

📄 PDF Abstract BibTeX arXiv:2306.03346

Code (1)

chongyi-zheng/stable_contrastive_rl 공식 구현 pytorch

Tasks

Contrastive Learning Data Augmentation Reinforcement Learning (RL)Self-Supervised Learning

Methods 이 논문이 사용한 방법론

Contrastive Learning 설명 없음

PRTS: A Primitive Reasoning and Tasking System via Contrastive Representations

2026-04-30 · Yang Zhang, Jiangyuan Zhao, Chenyou Fan, Fangzheng Yan 외 arxiv

Vision-Language-Action (VLA) models advance robotic control via strong visual-linguistic priors. However, existing VLAs predominantly frame pretraining as supervised behavior cloning, overlooking the fundamental nature o…

Reinforcement Learning

Offline Goal-conditioned Reinforcement Learning with Quasimetric Representations

2025-09-24 · Vivek Myers, Bill Chunyuan Zheng, Benjamin Eysenbach, Sergey Levine arxiv

Approaches for goal-conditioned reinforcement learning (GCRL) often use learned state representations to extract goal-reaching policies. Two frameworks for representation structure have yielded particularly effective GCR…

Representation Learning Reinforcement Learning Contrastive Learning

Learning Visual Robotic Control Efficiently with Contrastive Pre-training and Data Augmentation

2020-12-14 · Albert Zhan, Ruihan Zhao, Lerrel Pinto, Pieter Abbeel 외

Recent advances in unsupervised representation learning significantly improved the sample efficiency of training Reinforcement Learning policies in simulated environments. However, similar gains have not yet been seen fo…

Data Augmentation reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)+1

Leveraging Sequentiality in Reinforcement Learning from a Single Demonstration

2022-11-09 · Alexandre Chenu, Olivier Serris, Olivier Sigaud, Nicolas Perrin-Gilbert

Deep Reinforcement Learning has been successfully applied to learn robotic control. However, the corresponding algorithms struggle when applied to problems where the agent is only rewarded after achieving a complex task.…

Deep Reinforcement Learning reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Automaton Constrained Q-Learning

2025-10-06 · Anastasios Manganaris, Vittorio Giammarino, Ahmed H. Qureshi arxiv

Real-world robotic tasks often require agents to achieve sequences of goals while respecting time-varying safety constraints. However, standard Reinforcement Learning (RL) paradigms are fundamentally limited in these set…

Reinforcement Learning Continuous Control