Saved in:
Bibliographic Details
Main Authors: Zhang, Wenjing, Ding, Wandi, Zhu, Huaiping
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.16035
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912341158789120
author Zhang, Wenjing
Ding, Wandi
Zhu, Huaiping
author_facet Zhang, Wenjing
Ding, Wandi
Zhu, Huaiping
contents This paper highlights a parallel between the forward backward sweeping method for optimal control and deep learning training procedures. We reformulate a classical optimal control problem, constrained by a differential equation system, into an optimization framework that uses neural networks to represent control variables. We demonstrate that this deep learning method adheres to Pontryagin Maximum Principle and mitigates numerical instabilities by employing backward propagation instead of a backward sweep for the adjoint equations. As a case study, we solve an optimal control problem to find the optimal combination of immunotherapy and chemotherapy. Our approach holds significant potential across various fields, including epidemiology, ecological modeling, engineering, and financial mathematics, where optimal control under complex dynamic constraints is crucial.
format Preprint
id arxiv_https___arxiv_org_abs_2504_16035
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Universal differential equations for optimal control problems and its application on cancer therapy
Zhang, Wenjing
Ding, Wandi
Zhu, Huaiping
Optimization and Control
Dynamical Systems
34H05
This paper highlights a parallel between the forward backward sweeping method for optimal control and deep learning training procedures. We reformulate a classical optimal control problem, constrained by a differential equation system, into an optimization framework that uses neural networks to represent control variables. We demonstrate that this deep learning method adheres to Pontryagin Maximum Principle and mitigates numerical instabilities by employing backward propagation instead of a backward sweep for the adjoint equations. As a case study, we solve an optimal control problem to find the optimal combination of immunotherapy and chemotherapy. Our approach holds significant potential across various fields, including epidemiology, ecological modeling, engineering, and financial mathematics, where optimal control under complex dynamic constraints is crucial.
title Universal differential equations for optimal control problems and its application on cancer therapy
topic Optimization and Control
Dynamical Systems
34H05
url https://arxiv.org/abs/2504.16035