Salvato in:
Dettagli Bibliografici
Autore principale: Berger, Guillaume O.
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2511.20037
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911284944961536
author Berger, Guillaume O.
author_facet Berger, Guillaume O.
contents This paper studies the differentiability of the value function of switched linear systems under arbitrary switching and controlled switching, referred to as worst-case and optimal value functions respectively. First, we show that the value functions are Lipschitz continuous, when the cost function is Lipschitz continuous. Then, as the central contribution of this work, we show with examples that each of these functions can be non-differentiable on dense subsets of the state space, even if the cost function is smooth and Lipschitz continuous. This has implications for optimal control and reinforcement learning since it implies that the exact computation of these value functions requires templates involving functions that are non-differentiable on dense subsets.
format Preprint
id arxiv_https___arxiv_org_abs_2511_20037
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle On the differentiability of the value function of switched linear systems under arbitrary and controlled switching
Berger, Guillaume O.
Optimization and Control
This paper studies the differentiability of the value function of switched linear systems under arbitrary switching and controlled switching, referred to as worst-case and optimal value functions respectively. First, we show that the value functions are Lipschitz continuous, when the cost function is Lipschitz continuous. Then, as the central contribution of this work, we show with examples that each of these functions can be non-differentiable on dense subsets of the state space, even if the cost function is smooth and Lipschitz continuous. This has implications for optimal control and reinforcement learning since it implies that the exact computation of these value functions requires templates involving functions that are non-differentiable on dense subsets.
title On the differentiability of the value function of switched linear systems under arbitrary and controlled switching
topic Optimization and Control
url https://arxiv.org/abs/2511.20037