Self-Consistency of the Fokker-Planck Equation
Jul 2, 2022
Speakers
Zebang Shen
Sprecher:in · 0 Follower:innen
Zhenfu Wang
Sprecher:in · 0 Follower:innen
Satyen Kale
Sprecher:in · 0 Follower:innen
Alejandro Ribeiro
Sprecher:in · 0 Follower:innen
Amin Karbasi
Sprecher:in · 2 Follower:innen
Hamed Hassani
Sprecher:in · 3 Follower:innen
About
The Fokker-Planck equation (FPE) is the partial differential equation that governs the density evolution of the process and is of great importance to the literature of statistical physics and machine learning. The FPE can be regarded as a continuity equation where the change of the density is completely determined by a time varying velocity field. Importantly, this velocity field also depends on the current density function. As a result, the ground-truth velocity field can be shown to be the solution of a fixed-point equation, a property that we call self-consistency. In this paper, we exploit this concept to design a potential function of the hypothesis velocity fields, and prove that, if such a function diminishes to zero during the training procedure, the trajectory of the densities generated by the hypothesis velocity fields converges to the solution of the FPE in the Wasserstein-2 sense. The proposed potential function is amenable to neural-network based parameterization as the stochastic gradient with respect to the parameter can be efficiently computed. Once a parameterized model, such as Neural Ordinary Differential Equation is trained, we can generate the entire trajectory to the FPE.The Fokker-Planck equation (FPE) is the partial differential equation that governs the density evolution of the process and is of great importance to the literature of statistical physics and machine learning. The FPE can be regarded as a continuity equation where the change of the density is completely determined by a time varying velocity field. Importantly, this velocity field also depends on the current density function. As a result, the ground-truth velocity field can be shown to be the sol…
About COLT
The conference is held annually since 1988 and has become the leading conference on Learning theory by maintaining a highly selective process for submissions. It is committed in high-quality articles in all theoretical aspects of machine learning and related topics.
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
Gradient descent follows the regularization path for general losses
Später ansehen
Estimating Principal Components under Adversarial Perturbations
Später ansehen
Learning Entangled Single-Sample Gaussians in the Subset-of-Signals Model
Später ansehen
Hui Yuan, …
When Is Partially Observable Reinforcement Learning Not Scary?
Später ansehen
Qinghua Liu, …
