arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.07598cs.LG
收藏

通过最小框架平均实现更多对称性和效率

Equivariance via Minimal Frame Averaging for More Symmetries and Efficiency

Department of Computer Science and Engineering, Texas A\&M University, Texas, USA(计算机科学与工程系,德克萨斯农工大学,德克萨斯,美国)

Yuchao Lin, Jacob Helwig, Shurui Gui, Shuiwang Ji

AI总结:

本文提出最小框架平均框架,通过数学方法构建可证明最小的框架,实现精确的等变性,扩展到更多群,如洛伦兹群和酉群,并在多种任务中展示其高效性。

摘要:

我们考虑通过框架平均在机器学习系统中实现等变性。当前框架平均方法涉及对大框架的昂贵求和或依赖于采样方法,只能获得近似的等变性。这里,我们提出最小框架平均(MFA),一种用于构建可证明最小框架的数学框架,这些框架恰好是等变的。MFA的一般基础也使我们能够将框架平均扩展到之前考虑过的更多群,包括描述时空对称性的洛伦兹群和复值域的酉群。结果展示了通过MFA编码对称性在各种任务中的效率和有效性,包括n体模拟、对撞机物理中的顶标签识别和放松的能量预测。我们的代码可在https://github.com/divelab/MFA获得。

英文摘要:

We consider achieving equivariance in machine learning systems via frame averaging. Current frame averaging methods involve a costly sum over large frames or rely on sampling-based approaches that only yield approximate equivariance. Here, we propose Minimal Frame Averaging (MFA), a mathematical framework for constructing provably minimal frames that are exactly equivariant. The general foundations of MFA also allow us to extend frame averaging to more groups than previously considered, including the Lorentz group for describing symmetries in space-time, and the unitary group for complex-valued domains. Results demonstrate the efficiency and effectiveness of encoding symmetries via MFA across a diverse range of tasks, including $n$-body simulation, top tagging in collider physics, and relaxed energy prediction. Our code is available at https://github.com/divelab/MFA.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑