RESEARCH PAPERYear 2022

Contrastive Learning as Goal-Conditioned Reinforcement Learning

Benjamin Eysenbach; Tianjun Zhang; Sergey Levine; Russ R. Salakhutdinov

Classification

Major category: Foundational work
Quadrant: Not applicable
Architecture: Not applicable
Prediction paradigm: Not applicable
Subcategories: Theory & planning Action policy foundations
Source review status: Explicit in survey

Category review. Contrastive RL supplies a pre-2026 general relationship between future-state density ratios and goal-conditioned value learning, with a separate action-maximizing actor. This is a control-learning/theory foundation rather than a task-specific WAM. Reading evidence

AT A GLANCE

Contribution

A contribution summary has not been added yet.

Abstract

An abstract has not been added yet.

Affiliations

Not listed in the collection.

BibTeX

@article{eysenbach2022contrastive,
  title={Contrastive learning as goal-conditioned reinforcement learning},
  author={Eysenbach, Benjamin and Zhang, Tianjun and Levine, Sergey and Salakhutdinov, Russ R},
  journal={Advances in Neural Information Processing Systems},
  volume={35},
  pages={35603--35620},
  year={2022}
}

Classification

Contribution

Abstract

Affiliations

BibTeX

More in this category

Critique of World Model

Precise and Dexterous Robotic Manipulation via Human-in-the-Loop Reinforcement Learning

Mean Flows for One-step Generative Modeling