eess.SYSep 16, 2026

Learning and Control Beyond Linearity: Towards a Non-asymptotic Theory for Bilinear Systems

Authors: Yahya Sattar, Yassir Jedra, Robin Strässer, Frank Allgöwer, Maryam Fazel, Sarah Dean

Organizations: Cornell University · Imperial College London · University of Stuttgart · University of Washington Seattle · 

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

This tutorial provides a unified view of the emerging area of bilinear learning and control. Using linear systems as a benchmark, it explains what fundamentally changes in the bilinear settings, how recent theory addresses finite-sample learning and control, and how these ideas connect to broader themes in nonlinear control, representation learning, and data-driven decision making. For learning, we emphasize tools that are particularly useful in the bilinear settings, such as one-sided Bernstein's inequality for dependent and heavy-tailed covariates, blocking arguments, and martingale concentration for input-dependent noise. We then apply these tools to obtain finite-sample learning guarantees for fully observed bilinear systems, partially observed bilinear systems, and linear systems with bilinear observations. For control, we discuss quadratic control from bilinear observations, where the classical separation principle fails, and review tractable approaches based on belief-space receding horizon control. We also cover stabilization of bilinear dynamics under state feedback using semi-definite programming, LMI relaxations, sum-of-squares methods, and Koopman-based lifting. We conclude by discussing connections to reinforcement learning and machine learning, and some open problems in combined learning and control of bilinear systems.

More in this topic

Dynamical Systems

Dynamical systems studies how states evolve over time, including stability, feedback, chaos, and transitions in physical, biological, and engineered systems.

See all 478 papers

Explore similar work

CardsList
Apr 27, 2026

Dual Control of Linear Systems from Bilinear Observations with Belief Space Model Predictive Control

We study finite-horizon quadratic control of linear systems with bilinear observations, in which the control input affects not only the state dynamics but also the partial observations of the state. In this setting, the separation principle can fail because control inputs influence the future quality of state estimates. State estimation requires an input-dependent Kalman filter whose gain and error covariance evolve as functions of the control inputs. To address this challenge, we propose a belief-space model predictive control (B-MPC\texttt{B-MPC}) method that plans directly over both the estimated state and its error covariance. In particular, B-MPC\texttt{B-MPC} plans with a deterministic surrogate of the belief evolution defined by the input-dependent Kalman filter. Through numerical experiments in two synthetic settings, we show that B-MPC\texttt{B-MPC} can outperform both the separation-principle controller and its MPC variant in favorable regimes, and that these gains are accompanied by lower estimation covariance and more uncertainty-aware action choices.

Daniel Cao, Beixi Du, Andrew Lowitt +3

Cornell

Belief-Space PlanningModel Predictive Control
AskPDF
Sep 16, 2026

Demystifying Linear Operator Learning for Control Systems

This paper proposes a structured approach to learning linear operators for control systems from data. We address both structural and learning-theoretic aspects of the problem. To derive structural assumptions, we propose using the well-established framework of (semi)groups for evolution equations, as operators in control systems are of the same type. Further, we propose analyzing learning algorithms through the lens of the inverse problems framework. This reveals how a learned model depends on the data via error decompositions, convergence guarantees, and optimal regularization -- enabling us to compare existing methods and derive provably advantageous algorithms. In order to obtain these results, we restrict our scope to bounded operators on Hilbert spaces. Although this may appear restrictive, existing approaches often make this assumption implicitly to obtain matrix-like representations. We demonstrate the power of using these frameworks by deriving a convergent estimator for time-varying systems.

Max Beier, Nicolas Hoischen, Sandra Hirche +1

Chair of Information-Oriented Control, School of Computation, Information and Technology, Technical University of Munich, 80333 Munich, Germany

System IdentificationOperator Learning
AskPDF
Oct 8, 2026

Learning structured linear dynamical systems from missing observations

We consider the problem of learning structured linear dynamical systems over convex sets K\mathcal{K}, where only a small subset of the observations are available at each time point. An estimator which minimizes a bias-corrected, potentially non-convex objective function is proposed. Non-asymptotic bounds are obtained for the statistical error, which depend on the local complexity of K\mathcal{K}, the trajectory length TT, and the sub-sampling probability pp. Convergence of the projected gradient descent algorithm is also established. The general theory is applied to settings where (i) K\mathcal{K} is a subspace, (ii) K\mathcal{K} is the set of bi-isotonic matrices, and (iii) K\mathcal{K} is the set of matrices whose rows are formed by sampling Lipschitz functions. We show meaningful recovery of the transition matrix is possible for values of TT much smaller than what is required in the unconstrained case, and for p=o(1)p = o(1).

Aravinda Kanchana Ruwanpathirana, Hemant Tyagi, Sunny G. W. Wang

Division of Mathematical Sciences, SPMS, NTU Singapore 637371

System IdentificationLearning with Missing DataStochastic Linear Dynamical SystemsStatistical Learning Theory
AskPDF