将自监督的JEPA架构用于强化学习，解决模型坍缩问题。
JEPA for RL: Investigating Joint-Embedding Predictive Architectures for Reinforcement Learning
- 用联合嵌入预测架构构建视觉表征
- 在摆杆小车任务中实现稳定训练
- 适合研究自监督与强化学习结合
联合嵌入预测架构（JEPA）近年来成为自监督学习的有力候选。视觉变换器已通过JEPA从图像和视频中生成嵌入，展现出在分类、分割等下游任务中的优异性能。本文探讨如何将JEPA架构应用于基于图像的强化学习。我们分析了模型坍缩现象，提出有效预防方法，并在经典摆杆小车（Cart Pole）任务上提供了实证结果。
原文摘要 · Abstract (English)
Joint-Embedding Predictive Architectures (JEPA) have recently become popular as promising architectures for self-supervised learning. Vision transformers have been trained using JEPA to produce embeddings from images and videos, which have been shown to be highly suitable for downstream tasks like classification and segmentation. In this paper, we show how to adapt the JEPA architecture to reinforcement learning from images. We discuss model collapse, show how to prevent it, and provide exemplary data on the classical Cart Pole task.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。