RL with Self-Supervised 3D

Visual Reinforcement Learning with Self-Supervised 3D Representation

Visual Reinforcement Learning with Self-Supervised 3D Representations 1Shanghai Jiao Tong University, 2UC San Diego *Equal Contribution RA-L 2023 & IROS 2023 Abstract A prominent approach to visual Reinforcement Learning (RL) is to learn an internal state representation using self-supervised methods, which has the potential benefit of improved sample-efficiency and generalization through additional learning signal and inductive biases. However, while the real world is inherently 3D, prior efforts have largely been focused on leveraging 2D computer vision techniques as auxiliary…

saved by

• Fernando Silva

related reading

• [2103.06326] S4RL: Surprisingly Simple Self-Supervision for Offline Reinforcement Learningarxiv.org
• Precise Manipulation with Efficient Online RLpi.website
• [2506.09985] V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planningarxiv.org
• Learning 3D object-centric representation through predictionarxiv.org
• [2603.08546] Interactive World Simulator for Robot Policy Training and Evaluationarxiv.org
• SimpleVLA-RL: Scaling VLA Training via Reinforcement Learningalphaxiv.org
• Differentiable Robot Renderingdrrobot.cs.columbia.edu
• [2109.01115] Learning Language-Conditioned Robot Behavior from Offline Data and Crowd-Sourced Annotationarxiv.org
• e5b5c402bb7bd5e60bede6961d6fe39e-Paper-Conference.pdfproceedings.iclr.cc
• 2006.10742arxiv.org
• RL-VLM-F: Reinforcement Learning from Vision Language Foundation Model Feedbackarxiv.org
• Emergent Dexterity via Diverse Resets and Large-Scale Reinforcement Learningarxiv.org