arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.17211stat.MLcs.LG
收藏

MGD:基于矩引导的扩散用于最大熵生成

MGD: Moment Guided Diffusion for Maximum Entropy Generation

Département d’informatique, ENS, Université PSL(信息学院,巴黎高等师范学院,巴黎国立高等科学研究院);Capital Fund Management(资本基金管理公司);Collège de France(法兰西学院);Flatiron Institute(Flatiron研究所);Courant Institute of Mathematical Sciences, New York University(纽约大学应用数学科学学院);ML Lab, Capital Fund Management(资本基金管理公司的机器学习实验室)

Etienne Lempereur, Nathanaël Cuvelle--Magar, Florentin Coeurdoux, Stéphane Mallat, Eric Vanden-Eijnden

AI总结:

MGD结合最大熵方法与扩散模型,通过随机微分方程引导矩值,实现高效高维样本生成并提供理论保证。

摘要:

从有限信息中生成样本是跨科学领域的一个根本性问题。经典的最大熵方法通过矩约束提供原则性的不确定性量化,但需要通过MCMC或Langevin动力学进行采样,这在高维情况下通常会出现指数级的减速。相比之下,基于扩散和流匹配的生成模型能够高效地将噪声转化为数据,但提供有限的理论保证,并且在数据稀缺时容易过拟合。我们引入了矩引导扩散(MGD),结合了两种方法的元素。基于随机插值框架,MGD通过求解一个随机微分方程来采样最大熵分布,该方程在有限时间内引导矩值向预设值靠拢,从而避免了基于平衡方法的缓慢混合。我们正式地在大波动极限下,证明了MGD收敛于最大熵分布,并推导出从动力学直接计算所得熵的可计算估计器。应用到金融时间序列、湍流流体和宇宙学场中的小波散射矩,得到高维多尺度过程的负熵估计。

英文摘要:

Generating samples from limited information is a fundamental problem across scientific domains. Classical maximum entropy methods provide principled uncertainty quantification from moment constraints but require sampling via MCMC or Langevin dynamics, which typically exhibit exponential slowdown in high dimensions. In contrast, generative models based on diffusion and flow matching efficiently transport noise to data but offer limited theoretical guarantees and can overfit when data is scarce. We introduce Moment Guided Diffusion (MGD), which combines elements of both approaches. Building on the stochastic interpolant framework, MGD samples maximum entropy distributions by solving a stochastic differential equation that guides moments toward prescribed values in finite time, thereby avoiding slow mixing in equilibrium-based methods. We formally obtain, in the large-volatility limit, convergence of MGD to the maximum entropy distribution and derive a tractable estimator of the resulting entropy computed directly from the dynamics. Applications to financial time series, turbulent flows, and cosmological fields using wavelet scattering moments yield estimates of negentropy for high-dimensional multiscale processes.

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。

↑