Salvato in:
Dettagli Bibliografici
Autori principali: Hua, Mengjian, Laurière, Mathieu, Vanden-Eijnden, Eric
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2410.05163
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916733846028288
author Hua, Mengjian
Laurière, Mathieu
Vanden-Eijnden, Eric
author_facet Hua, Mengjian
Laurière, Mathieu
Vanden-Eijnden, Eric
contents We present a novel on-policy algorithm for solving stochastic optimal control (SOC) problems. By leveraging the Girsanov theorem, our method directly computes on-policy gradients of the SOC objective without expensive backpropagation through stochastic differential equations or adjoint problem solutions. This approach significantly accelerates the optimization of neural network control policies while scaling efficiently to high-dimensional problems and long time horizons. We evaluate our method on classical SOC benchmarks as well as applications to sampling from unnormalized distributions via Schrödinger-Föllmer processes and fine-tuning pre-trained diffusion models. Experimental results demonstrate substantial improvements in both computational speed and memory efficiency compared to existing approaches.
format Preprint
id arxiv_https___arxiv_org_abs_2410_05163
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle An Efficient On-Policy Deep Learning Framework for Stochastic Optimal Control
Hua, Mengjian
Laurière, Mathieu
Vanden-Eijnden, Eric
Machine Learning
Optimization and Control
We present a novel on-policy algorithm for solving stochastic optimal control (SOC) problems. By leveraging the Girsanov theorem, our method directly computes on-policy gradients of the SOC objective without expensive backpropagation through stochastic differential equations or adjoint problem solutions. This approach significantly accelerates the optimization of neural network control policies while scaling efficiently to high-dimensional problems and long time horizons. We evaluate our method on classical SOC benchmarks as well as applications to sampling from unnormalized distributions via Schrödinger-Föllmer processes and fine-tuning pre-trained diffusion models. Experimental results demonstrate substantial improvements in both computational speed and memory efficiency compared to existing approaches.
title An Efficient On-Policy Deep Learning Framework for Stochastic Optimal Control
topic Machine Learning
Optimization and Control
url https://arxiv.org/abs/2410.05163