Statistical learning theory for control: A finite-sample perspective

A Tsiamis, I Ziemann, N Matni… - IEEE Control Systems …, 2023 - ieeexplore.ieee.org
Learning algorithms have become an integral component to modern engineering solutions.
Examples range from self-driving cars and recommender systems to finance and even …

Neural operators for bypassing gain and control computations in pde backstepping

L Bhan, Y Shi, M Krstic - IEEE Transactions on Automatic …, 2023 - ieeexplore.ieee.org
We introduce a framework for eliminating the computation of controller gain functions in PDE
control. We learn the nonlinear operator from the plant parameters to the control gains with a …

Maximum diffusion reinforcement learning

TA Berrueta, A Pinosky, TD Murphey - Nature Machine Intelligence, 2024 - nature.com
Robots and animals both experience the world through their bodies and senses. Their
embodiment constrains their experiences, ensuring that they unfold continuously in space …

The power of learned locally linear models for nonlinear policy optimization

D Pfrommer, M Simchowitz… - International …, 2023 - proceedings.mlr.press
A common pipeline in learning-based control is to iteratively estimate a model of system
dynamics, and apply a trajectory optimization algorithm-eg $\mathtt {iLQR} $-on the learned …

Meta-learning operators to optimality from multi-task non-iid data

TTCK Zhang, LF Toso, J Anderson, N Matni - arXiv preprint arXiv …, 2023 - arxiv.org
A powerful concept behind much of the recent progress in machine learning is the extraction
of common features across data from heterogeneous sources or tasks. Intuitively, using all of …

Regret lower bounds for learning linear quadratic gaussian systems

I Ziemann, H Sandberg - IEEE Transactions on Automatic …, 2024 - ieeexplore.ieee.org
In this article, we establish regret lower bounds for adaptively controlling an unknown linear
Gaussian system with quadratic costs. We combine ideas from experiment design …

How are policy gradient methods affected by the limits of control?

I Ziemann, A Tsiamis, H Sandberg… - 2022 IEEE 61st …, 2022 - ieeexplore.ieee.org
We study stochastic policy gradient methods from the perspective of control-theoretic
limitations. Our main result is that ill-conditioned linear systems in the sense of Doyle …

Sample-Efficient Linear Representation Learning from Non-IID Non-Isotropic Data

TTCK Zhang, LF Toso, J Anderson… - The Twelfth International …, 2024 - openreview.net
A powerful concept behind much of the recent progress in machine learning is the extraction
of common features across data from heterogeneous sources or tasks. Intuitively, using all of …

End-to-end guarantees for indirect data-driven control of bilinear systems with finite stochastic data

N Chatzikiriakos, R Strässer, F Allgöwer… - arXiv preprint arXiv …, 2024 - arxiv.org
In this paper we propose an end-to-end algorithm for indirect data-driven control for bilinear
systems with stability guarantees. We consider the case where the collected iid data is …

The fundamental limitations of learning linear-quadratic regulators

BD Lee, I Ziemann, A Tsiamis… - 2023 62nd IEEE …, 2023 - ieeexplore.ieee.org
We present a local minimax lower bound on the excess cost of designing a linear-quadratic
controller from offline data. The bound is valid for any offline exploration policy that consists …