Enregistré dans:
Détails bibliographiques
Auteurs principaux: Makdah, Abed AlRahman Al, Kosut, Oliver, Sankar, Lalitha, Zou, Shaofeng
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2508.17185
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866914002038161408
author Makdah, Abed AlRahman Al
Kosut, Oliver
Sankar, Lalitha
Zou, Shaofeng
author_facet Makdah, Abed AlRahman Al
Kosut, Oliver
Sankar, Lalitha
Zou, Shaofeng
contents Many applications -- including power systems, robotics, and economics -- involve a dynamical system interacting with a stochastic and hard-to-model environment. We adopt a reinforcement learning approach to control such systems. Specifically, we consider a deterministic, discrete-time, linear, time-invariant dynamical system coupled with a feature-based linear Markov process with an unknown transition kernel. The objective is to learn a control policy that optimizes a quadratic cost over the system state, the Markov process, and the control input. Leveraging both components of the system, we derive an explicit parametric form for the optimal state-action value function and the corresponding optimal policy. Our model is distinct in combining aspects of both classical Linear Quadratic Regulator (LQR) and linear Markov decision process (MDP) frameworks. This combination retains the implementation simplicity of LQR, while allowing for sophisticated stochastic modeling afforded by linear MDPs, without estimating the transition probabilities, thereby enabling direct policy improvement. We use tools from control theory to provide theoretical guarantees on the stability of the system under the learned policy and provide a sample complexity analysis for its convergence to the optimal policy. We illustrate our results via a numerical example that demonstrates the effectiveness of our approach in learning the optimal control policy under partially known stochastic dynamics.
format Preprint
id arxiv_https___arxiv_org_abs_2508_17185
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Linear Dynamics meets Linear MDPs: Closed-Form Optimal Policies via Reinforcement Learning
Makdah, Abed AlRahman Al
Kosut, Oliver
Sankar, Lalitha
Zou, Shaofeng
Optimization and Control
Systems and Control
Many applications -- including power systems, robotics, and economics -- involve a dynamical system interacting with a stochastic and hard-to-model environment. We adopt a reinforcement learning approach to control such systems. Specifically, we consider a deterministic, discrete-time, linear, time-invariant dynamical system coupled with a feature-based linear Markov process with an unknown transition kernel. The objective is to learn a control policy that optimizes a quadratic cost over the system state, the Markov process, and the control input. Leveraging both components of the system, we derive an explicit parametric form for the optimal state-action value function and the corresponding optimal policy. Our model is distinct in combining aspects of both classical Linear Quadratic Regulator (LQR) and linear Markov decision process (MDP) frameworks. This combination retains the implementation simplicity of LQR, while allowing for sophisticated stochastic modeling afforded by linear MDPs, without estimating the transition probabilities, thereby enabling direct policy improvement. We use tools from control theory to provide theoretical guarantees on the stability of the system under the learned policy and provide a sample complexity analysis for its convergence to the optimal policy. We illustrate our results via a numerical example that demonstrates the effectiveness of our approach in learning the optimal control policy under partially known stochastic dynamics.
title Linear Dynamics meets Linear MDPs: Closed-Form Optimal Policies via Reinforcement Learning
topic Optimization and Control
Systems and Control
url https://arxiv.org/abs/2508.17185