世界模型和VLA学习记录-小白（探索学习中并会持续更新，VLA未开始学）

前言作为智能驾驶领域的感知测试人员目前只接触到了感知评测方案是传统的感知-预测-规划-控制模式。现在随着AI大模型的发展个人认为智驾未来的发展方向是端到端大模型并且我对世界模型和VLA比较感兴趣因此写这篇笔记去记录学习过程。世界模型1. 了解发展过程世界模型 (World Models) 必读论文_世界模型论文-CSDN博客1.1 阅读RNN规划的奠基性论文Jürgen Schmidhuber在1990年发表的技术报告《Making the World Differentiable: On Using Self-Supervised Fully Recurrent Neural Networks for Dynamic Reinforcement Learning and Planning in Non-Stationary Environments》简称FKI-126-90报告1.2 阅读Sutton (1990) - Dyna架构论文笔记Dyna, an Integrated Architecture for Learning, Planning, and Reacting_dyna论文详解-CSDN博客1.3 阅读《World Models》 (Ha Schmidhuber, 2018论文阅读_世界模型_world model论文-CSDN博客1.4 阅读PlaNet (Hafner et al., 2018)强化学习论文(5): Learning Latent Dynamics for Planning from Pixels-CSDN博客【论文笔记】Learning Latent Dynamics for Planning from Pixels-CSDN博客1.5 阅读Dreamer系列Mastering diverse control tasks through world models论文精读-CSDN博客1.6 阅读MuZero (DeepMind, 2020)‌《Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model》https://zhuanlan.zhihu.com/p/220695590181.7 阅读LeCun的白皮书 (2022)A Path Towards Autonomous Machine Intelligence Version文章精读-CSDN博客1.8 阅读V-JEPA (Meta, 2024)JEPA-VLA视频预测嵌入如何革新机器人视觉-语言-动作模型_vla-jepa模型论文-CSDN博客V-JEPA1论文下载链接[2404.08471] Revisiting Feature Prediction for Learning Visual Representations from VideoV-JEPA2论文下载链接https://ai.meta.com/research/publications/v-jepa-2-self-supervised-video-models-enable-understanding-prediction-and-planning/