arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2604.20188cs.LGmath.DS
收藏

具有结构意识的变分学习一类广义扩散过程

Structure-Aware Variational Learning of a Class of Generalized Diffusions

School of Mathematics, South China University of Technology(华南理工大学数学学院);Department of Applied Mathematics, Illinois Institute of Technology(伊利诺伊理工学院应用数学系);School of Computational Science and Engineering, Georgia Institute of Technology(佐治亚理工学院计算科学与工程学院)

Yubin Lu, Xiaofan Li, Chun Liu, Qi Tang, Yiwei Wang

AI总结:

本文提出基于能量的变分学习框架,通过能量耗散原理从部分和噪声观测中学习随机梯度系统的潜在能量函数,提升对观测时间、噪声水平和数据多样性的鲁棒性。

摘要:

从随机梯度系统的潜在能量学习是一个在物理、化学和数据驱动建模中出现的基础问题。经典方法常依赖于直接回归控制方程或速度场,容易受噪声和外部扰动影响,当观测不完整时可能失效。本文提出一种结构意识、基于能量的学习框架,用于推断广义扩散过程中的未知势函数,基于能量变分方法。从与福克-计划克方程相关的能量耗散定律出发,基于德吉奥尼耗散泛函构造损失函数,一致地耦合系统的自由能和耗散机制。该方法避免了显式执行控制偏微分方程,保留了动力学的变分结构。通过一、二、三维的数值实验,证明所提出的基于能量的损失函数在观测时间、噪声水平和可用训练数据的多样性及数量方面具有增强的鲁棒性。这些结果突显了能量耗散原理作为从数据中学习随机扩散动力学可靠基础的有效性。

英文摘要:

Learning the underlying potential energy of stochastic gradient systems from partial and noisy observations is a fundamental problem arising in physics, chemistry, and data-driven modeling. Classical approaches often rely on direct regression of governing equations or velocity fields, which can be sensitive to noise and external perturbations and may fail when observations are incomplete. In this work, we propose a structure-aware, energy-based learning framework for inferring unknown potential functions in generalized diffusion processes, grounded in the energetic variational approach. Starting from the energy-dissipation law associated with the Fokker-Planck equation, we construct loss functions based on the De Giorgi dissipation functional, which consistently couple the free energy and the dissipation mechanism of the system. This formulation avoids explicit enforcement of the governing partial differential equation and preserves the underlying variational structure of the dynamics. Through numerical experiments in one, two, and three dimensions, we demonstrate that the proposed energy-based loss exhibits enhanced robustness with respect to observation time, noise level, and the diversity and amount of available training data. These results highlight the effectiveness of energy-dissipation principles as a reliable foundation for learning stochastic diffusion dynamics from data.

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。

↑