潜变量能量模型导论:迈向自主机器智能之路
机器学习
2024-10-31 v1 无序系统与神经网络
机器学习
摘要
当前自动化系统存在关键局限,在人工智能达到类人水平并带来新技术革命之前亟需解决。其中,我们的社会仍缺乏 Level 5 自动驾驶汽车、家用机器人与可学习可靠世界模型、推理并规划复杂动作序列的虚拟助手。在本笔记中,我们总结了 Yann LeCun 所提出的未来自主智能架构背后的主要思想。具体而言,我们引入能量模型与潜变量模型,并将其优势结合于 LeCun 提案的构建模块中,即分层联合嵌入预测架构(H-JEPA)。
关键词
引用
@article{arxiv.2306.02572,
title = {Introduction to Latent Variable Energy-Based Models: A Path Towards Autonomous Machine Intelligence},
author = {Anna Dawid and Yann LeCun},
journal= {arXiv preprint arXiv:2306.02572},
year = {2024}
}
备注
23 pages + 1-page appendix, 11 figures. These notes follow the content of three lectures given by Yann LeCun during the Les Houches Summer School on Statistical Physics and Machine Learning in 2022. Feedback and comments are most welcome!