paper-with-me

Papers

PAC-Bayesian Certificates for Quadratic Closed-Loop Control

2026-06-26 · Domagoj Herceg arxiv

PAC-Bayesian bounds provide finite-sample guarantees for data-dependent randomized predictors, but applying them to learning-based control is difficult because the natural objective is a quadratic trajectory cost. Such losses are unbounded, non-Lipschitz , and lead to response-dependent Chernoff terms. We employ System Level Synthesis parameterization, which exposes the closed-loop trajectory map of a linear system directly and makes the quadratic control loss amenable to explicit certification. Moreover, we provide a set of PAC-Bayes-Chernoff certificates for posterior distributions over feasible closed-loop responses. For Gaussian disturbance trajectories with arbitrary covariance, we derive an exact one-sided Gaussian transform and a tractable quadratic upper bound expressed through closed-loop sensitivity quantities. We also derive a posterior-localized surrogate for settings where pointwise closed-loop response certificates are unavailable or have support related admissibility issues. Although PAC-Bayes certifies a non-degenerate posterior, the convex quadratic form of the SLS loss transfers the certificate to the posterior mean response. We present a deterministic mean response deployment result that is particularly suitable for control while retaining the stochastic posterior in the bound. Additionally, we provide a data-driven bound for this deployment, transitioning away from an oracle bound. Minimizing this bound naturally results in a learning algorithm for control selection from data. Numerical experiments on a double integrator show that the algorithm acts as a sensitivity-aware finite-sample regularizer, improving held-out cost and reducing closed-loop sensitivity in the low-data regime

📄 PDF Abstract BibTeX arXiv:2606.28281

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

MR-ARL: Model Reference Adaptive Reinforcement Learning for Robustly Stable On-Policy Data-Driven LQR

2024-02-22 · Marco Borghesi, Alessandro Bosso, Giuseppe Notarstefano

This article introduces a novel framework for data-driven linear quadratic regulator (LQR) design. First, we introduce a reinforcement learning paradigm for on-policy data-driven LQR, where exploration and exploitation a…

reinforcement-learningReinforcement Learning

Distributionally Robust PAC-Bayesian Control

2026-04-12 · Domagoj Herceg, Duarte Antunes arxiv

We present a distributionally robust PAC-Bayesian framework for certifying the performance of learning-based finite-horizon controllers. While existing PAC-Bayes control literature typically assumes bounded losses and ma…

A relaxed technical assumption for posterior sampling-based reinforcement learning for control of unknown linear systems

2021-08-19 · Mukul Gagrani, Sagar Sudhakara, Aditya Mahajan, Ashutosh Nayyar 외

We revisit the Thompson sampling algorithm to control an unknown linear quadratic (LQ) system recently proposed by Ouyang et al (arXiv:1709.04047). The regret bound of the algorithm was derived under a technical assumpti…

Thompson Sampling

Data-Driven Controller Design via Finite-Horizon Dissipativity

2021-01-15 · Nils Wieler, Julian Berberich, Anne Koch, Frank Allgöwer

Given one open-loop measured trajectory of a single-input single-output discrete-time linear time-invariant system, we present a framework for data-driven controller design for closed-loop finite-horizon dissipativity. F…

High-Dimensional Surrogate Modeling for Closed-Loop Learning of Neural-Network-Parameterized Model Predictive Control

2025-12-12 · Sebastian Hirt, Valentinus Suwanto, Hendrik Alsmeier, Maik Pfefferkorn 외 arxiv

Learning controller parameters from closed-loop data has been shown to improve closed-loop performance. Bayesian optimization, a widely used black-box and sample-efficient learning method, constructs a probabilistic surr…

Gaussian Processes