arXiv subjects

Andrea Iannelli

Publications and source records attributed to Andrea Iannelli.

At least 19 recordsLinked to original sources

arXiv2026

Structure, Analysis, and Synthesis of First-Order Algorithms

First-order optimization algorithms can be interpreted through the lens of dynamical systems as the interconnection of linear systems and a set of subgradient nonlinearities. This dynamical systems formulation allows for the analysis and synthesis of optimization algorithms by solving robust control problems. In this work, we use the celebrated internal model principle in control theory to structurally factorize convergent composite optimization algorithms into suitable internal models and core subcontrollers. As the key benefit, we reveal that this permits us to synthesize optimization algorithms even if information is transmitted over networks featuring dynamical phenomena such as time delays, channel memory, or crosstalk. Design of certified linearly-convergent algorithms is achieved under bisection either through a nonconvex local search or by alternation of convex semidefinite programs. We demonstrate factorization of existing optimization algorithms and the automated synthesis of new optimization algorithms in the networked setting.

math.OC↗

arXiv2026

Sample-Efficient Model-Free Policy Gradient Methods for Stochastic LQR via Robust Linear Regression

Policy gradient algorithms are widely used in reinforcement learning and belong to the class of approximate dynamic programming methods. This paper studies two key policy gradient algorithms, the Natural Policy Gradient and the Gauss-Newton Method, for solving the Linear Quadratic Regulator (LQR) problem in unknown stochastic linear systems. The main challenge lies in obtaining a consistent gradient estimate from noisy data due to errors-in-variables in linear regression. This issue is addressed by employing a primal-dual estimation procedure. Using this novel gradient estimation scheme, the paper establishes convergence guarantees with a sample complexity of order O(1/epsilon). Theoretical results are further supported by numerical experiments, which demonstrate the effectiveness of the proposed algorithms.

eess.SY↗

arXiv2026

Adaptive Stepsizes With Certified Convergence in Distributed Gradient Tracking With Quadratic Costs

In this work, we propose an adaptive stepsize rule with guaranteed convergence for Distributed Gradient Tracking applied to scalar quadratic problems with heterogeneous curvatures. Most distributed gradient-based algorithms require a suitable stepsize selection. Available theoretical bounds are often overly conservative, while practical implementations typically rely on empirically tuned heuristics. Online adaptive strategies have only recently emerged for general distributed convex optimization, but their properties and performance remain only partially understood. To gain analytical insight, we focus on the informative setting of scalar quadratic costs, which allows us to explicitly capture the interplay between network topology and curvature heterogeneity. We derive a convergence bound parameterized only by the essential spectral radius of the consensus matrix and the heterogeneity of the local cost curvatures, both computable online without any prior knowledge of the optimization problem. Optimizing this bound yields a computationally tractable surrogate for the convergence rate and the optimal constant stepsize. The resulting stepsize admits an analytical interpretation, guarantees convergence for arbitrary network topologies and curvature heterogeneity, and is provably tight for complete graphs and homogeneous curvatures. Finally, extensive numerical simulations demonstrate that the proposed distributed adaptive strategy significantly outperforms existing offline and online stepsize selection rules in the considered setting.

math.OC↗

arXiv2026

Accelerated ADMM: Automated Parameter Tuning and Improved Linear Convergence

This work studies the linear convergence of an accelerated scheme of the Alternating Direction Method of Multipliers (ADMM) for strongly convex and Lipschitz-smooth problems. We use the methodology of expressing the accelerated ADMM as a Lur'e system, i.e., an interconnection of a linear dynamical system in feedback with a slope-restricted operator, and we use Integral Quadratic Constraints to establish linear convergence. In addition, we propose several parameter tuning heuristics and their impact on the convergence rate through numerical analyses. Our new bounds show improved linear convergence rates compared to the vanilla algorithm and previous proposed accelerated variants, which is also empirically validated on a LASSO regression benchmark.

math.OC↗

arXiv2026

A Generalized Plant Perspective on Linear-Convex Feedback Optimization

Feedback optimization is a control approach for driving a dynamical system to the solution of an optimization problem by interconnecting the plant with an algorithm. Existing stability guarantees typically rely on timescale separation, enforced by conservative gain bounds that limit transient performance and require a pre-stabilized plant. This paper revisits the robust control perspective on feedback optimization. We formulate the plant-optimizer interconnection as a generalized plant, where the cost gradients are characterized by Zames--Falb Integral Quadratic Constraints. Classical timescale-separation bounds are recovered as a special case of static multipliers, with dynamic multipliers yielding substantially tighter stability margins. The formulation also enables IQC based synthesis of dynamic output feedback controllers that jointly stabilize the plant and optimize transient performance, with possible model uncertainty absorbed into an uncertainty channel. For constrained problems, the framework extends to dynamic controllers that generalize projected gradient flows. Numerical examples illustrate the benefits and flexibility of the proposed approach.

eess.SY↗

arXiv2026

Beyond Bounded Noise: Stochastic Set-Membership Estimation for Nonlinear Systems

In this paper, we derive a novel procedure for set-membership estimation of dynamical systems affected by stochastic noise with unbounded support. Employing a bound on the sample covariance matrix, we are able to provide a finite- sample uncertainty set containing the true system parameters with high probability. Our approach can be natively applied to a wide class of nonlinear systems affected by sub-Gaussian noise. Our analysis provides conditions under which the proposed uncertainty set converges to the true system parameters and establishes an upper bound on the convergence rate. The proposed uncertainty set can be used directly for robust controller synthesis with probabilistic stability and performance guarantees. Concluding numerical examples demonstrate the advantages of the proposed formulation over established approaches.

eess.SY↗

arXiv2026

A Unified Bayesian Framework for Data-Driven Smoothing, Prediction, and Control

Extending data-driven algorithms based on Willems' fundamental lemma to stochastic data often requires empirical and customized workarounds. This work presents a unified Bayesian framework for linear systems that provides a systematic and general method for handling stochastic data-driven tasks, including smoothing, prediction, and control, via maximum a posteriori estimation. This framework formulates a unified trajectory estimation problem for the three tasks by specifying different types of trajectory knowledge. Then, a Bayesian problem is solved that optimally combines trajectory knowledge with a data-driven characterization of the trajectory from offline data for correlated input-output uncertainties with elliptical distributions. Under specific conditions, this problem is shown to generalize existing data-driven prediction and control algorithms. Numerical examples demonstrate the performance of the unified approach for all three tasks against other data-driven and system identification approaches.

eess.SY↗

arXiv2026

Convergence Guarantees of Model-free Policy Gradient Methods for LQR with Stochastic Data

Policy gradient (PG) methods are the backbone of many reinforcement learning algorithms due to their good performance in policy optimization problems. As a gradient-based approach, PG methods typically rely on knowledge of the system dynamics. If this is not available, trajectory data can be utilized to approximate first-order information. When the data are noisy, gradient estimates become inaccurate and a study that investigates uncertainty estimation and the analysis of its propagation through the algorithm is currently missing. To address this, our work focuses on the Linear Quadratic Regulator (LQR) problem for systems subject to additive stochastic noise. After briefly summarizing the state of the art for cases with a known model, we focus on scenarios where the system dynamics are unknown, and approximate gradient information is obtained using zeroth-order optimization techniques. We analyze the theoretical properties by computing the error in the estimated gradient and examining how this error affects the convergence of PG algorithms. Additionally, we provide global convergence guarantees for various versions of PG methods, including those employing adaptive step sizes and variance reduction techniques, which help increase the convergence rate and reduce sample complexity. This study contributed to characterizing the robustness of model-free PG methods, aiming to identify their limitations in the presence of stochastic noise and proposing improvements to enhance their applicability.

eess.SY↗

arXiv2026

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

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 i.i.d. data is affected by probabilistic noise with possibly unbounded support and leverage tools from statistical learning theory to derive finite sample identification error bounds. To this end, we solve the bilinear identification problem by solving a set of linear and affine identification problems, by a particular choice of a control input during the data collection phase. We provide a priori as well as data-dependent finite sample identification error bounds on the individual matrices as well as ellipsoidal bounds, both of which are structurally suitable for control. Further, we integrate the structure of the derived identification error bounds in a robust controller design to obtain an exponentially stable closed-loop. By means of an extensive numerical study we showcase the interplay between the controller design and the derived identification error bounds. Moreover, we note appealing connections of our results to indirect data-driven control of general nonlinear systems through Koopman operator theory and discuss how our results may be applied in this setup.

eess.SY↗

arXiv2026

Analysis and Synthesis of Switched Optimization Algorithms

Deployment of optimization algorithms over communication networks face challenges associated with time delays and corruptions. Fixed time delays can destabilize popular gradient-based algorithms, and this degradation is exacerbated by time-varying delays that may arise from packet drops. This work concentrates on the analysis and synthesis of discrete-time optimization algorithms with certified exponential convergence rates that are robust against switched network dynamics between the optimizer and the gradient oracle. Analysis is accomplished by solving linear matrix inequalities under bisection in the exponential convergence rate, searching over Zames-Falb filter coefficients that can certify convergence. Synthesis is performed by alternating between a search over filter coefficient for a fixed controller, and a search over controllers for a fixed filter. Effectiveness is demonstrated by the synthesis of convergent optimization algorithms over networks with time-varying delays, and networks with unstable channel dynamics.

math.OC↗

arXiv2026

A Linear Parameter-Varying Framework for the Analysis of Time-Varying Optimization Algorithms

In this paper we propose a framework to analyze iterative first-order optimization algorithms for time-varying convex optimization. We assume that the temporal variability is caused by a time-varying parameter entering the objective, which can be measured at the time of decision but whose future values are unknown. We consider the case of strongly convex objective functions with Lipschitz continuous gradients under a convex constraint set. We model the algorithms as discrete-time linear parameter varying (LPV) systems in feedback with monotone operators such as the time-varying gradient. We leverage the approach of analyzing algorithms as uncertain control interconnections with integral quadratic constraints (IQCs) and generalize that framework to the time-varying case. We propose novel IQCs that are capable of capturing the behavior of time-varying nonlinearities and leverage techniques from the LPV literature to establish novel bounds on the tracking error. Quantitative bounds can be computed by solving a semi-definite program and can be interpreted as an input-to-state stability result with respect to a disturbance signal which increases with the temporal variability of the problem. As a departure from results in this research area, our bounds introduce a dependence on different additional measures of temporal variations, such as the function value and gradient rate of change. We exemplify our main results with numerical experiments that showcase how our analysis framework is able to capture convergence rates of different first-order algorithms for time-varying optimization through the choice of IQC and rate bounds.

math.OC↗

arXiv2026

A Stochastic Gradient Descent Approach to Design Policy Gradient Methods for LQR

In this work, we propose a stochastic gradient descent (SGD) framework to design data-driven policy gradient descent algorithms for the linear quadratic regulator problem. Two alternative schemes are considered to estimate the policy gradient from stochastic trajectory data: (i) an indirect online identification based approach, in which the system matrices are first estimated and subsequently used to construct the gradient, and (ii) a direct zeroth-order approach, which approximates the gradient using empirical cost evaluations. In both cases, the resulting gradient estimates are random due to stochasticity in the data, allowing us to use SGD theory to analyze the convergence of the associated policy gradient methods. A key technical step consists of modeling the gradient estimates as suitable stochastic gradient oracles, which, because of the way they are computed, are inherently based. We derive sufficient conditions under which SGD with a biased gradient oracle converges asymptotically to the optimal policy, and leverage these conditions to design the parameters of the gradient estimation schemes. Moreover, we compare the advantages and limitations of the two data-driven gradient estimators. Numerical experiments validate the effectiveness of the proposed methods.

eess.SY↗

arXiv2026

High Effort, Low Gain: Fundamental Limits of Active Learning for Linear Dynamical Systems

In this work, we consider the problem of identifying an unknown linear dynamical system given a finite hypothesis class. In particular, we analyze the effect of the excitation input on the sample complexity of identifying the true system with high probability. To this end, we present sample complexity lower bounds that capture the choice of the selected excitation input. The sample complexity lower bound gives rise to a system theoretic condition to determine the potential benefit of experiment design. Informed by the analysis of the sample complexity lower bound, we propose a persistent excitation (PE) condition tailored to the considered setting, which we then use to establish sample complexity upper bounds. Notably, the PE condition is weaker than in the case of an infinite hypothesis class and allows analyzing different excitation inputs modularly. Crucially, the lower and upper bounds share the same dependency on key problem parameters. Finally, we leverage these insights to propose an active learning algorithm that sequentially excites the system optimally with respect to the current estimate, and provide sample complexity guarantees for the presented algorithm. Concluding simulations showcase the effectiveness of the proposed algorithm.

eess.SY↗

arXiv2025

Adaptive Time-Domain Harmonic Control for Noise-Vibration-Harshness Reduction of Electric Drives

Reducing Noise, Vibration, and Harshness (NVH) in electric drives is crucial for applications such as electric vehicle drivetrains and heat-pump compressors, where strict NVH requirements directly affect user satisfaction and component longevity. This work presents the integration of an adaptive time-domain harmonic controller into an existing electric-drive control loop to attenuate harmonic disturbances. Three control structures are proposed and analyzed, along with a modified parameter-estimation scheme that reduces computational effort while preserving estimation accuracy, making the method suitable for embedded real-time implementation. To cope with fast operating-point changes, a delta-learning approach combines adaptive control with a lookup-table-based feedforward estimator, ensuring fast convergence and robustness. The proposed controller architectures are validated through simulation and testbench experiments on a permanent-magnet synchronous machine drive, demonstrating substantial NVH reductions across operating conditions. The results confirm that time-domain adaptive harmonic control offers a practical and theoretically grounded solution for real-time NVH mitigation in electric drives.

eess.SY↗

arXiv2025

Convergence and stability of Q-learning in Hierarchical Reinforcement Learning

Hierarchical Reinforcement Learning promises, among other benefits, to efficiently capture and utilize the temporal structure of a decision-making problem and to enhance continual learning capabilities, but theoretical guarantees lag behind practice. In this paper, we propose a Feudal Q-learning scheme and investigate under which conditions its coupled updates converge and are stable. By leveraging the theory of Stochastic Approximation and the ODE method, we present a theorem stating the convergence and stability properties of Feudal Q-learning. This provides a principled convergence and stability analysis tailored to Feudal RL. Moreover, we show that the updates converge to a point that can be interpreted as an equilibrium of a suitably defined game, opening the door to game-theoretic approaches to Hierarchical RL. Lastly, experiments based on the Feudal Q-learning algorithm support the outcomes anticipated by theory.

cs.LG↗

arXiv2025

Data-Driven Stabilization of Continuous-Time LTI Systems from Noisy Input-Output Data

We present an approach to compute stabilizing controllers for continuous-time linear time-invariant systems directly from an input-output trajectory affected by process and measurement noise. The proposed output-feedback design combines (i) an observer of a non-minimal realization of the plant and (ii) a feedback law obtained from a linear matrix inequality (LMI) that depends solely on the available data. Under a suitable interval excitation condition and knowledge of a noise energy bound, the feasibility of the LMI is shown to be necessary and sufficient for stabilizing all non-minimal realizations consistent with the data. We further provide a condition for the feasibility of the LMI related to the signal-to-noise ratio, guidelines to compute the noise energy bound, and numerical simulations that illustrate the effectiveness of the approach.

eess.SY↗

arXiv2025

A polynomial-based QCQP solver for encrypted optimization

In this paper, we present a novel method for solving a class of quadratically constrained quadratic optimization problems using only additions and multiplications. This approach enables solving constrained optimization problems on private data since the operations involved are compatible with the capabilities of homomorphic encryption schemes. To solve the constrained optimization problem, a sequence of polynomial penalty functions of increasing degree is introduced, which are sufficiently steep at the boundary of the feasible set. Adding the penalty function to the original cost function creates a sequence of unconstrained optimization problems whose minimizer always lies in the admissible set and converges to the minimizer of the constrained problem. A gradient descent method is used to generate a sequence of iterates associated with these problems. For the algorithm, it is shown that the iterate converges to a minimizer of the original problem, and the feasible set is positively invariant under the iteration. Finally, the method is demonstrated on an illustrative cryptographic problem, finding the smaller value of two numbers, and the encrypted implementability is discussed.

math.OC↗

arXiv2025

Hidden Convexity in Active Learning: A Convexified Online Input Design for ARX Systems

The goal of this work is to accelerate the identification of an unknown ARX system from trajectory data through online input design. Specifically, we present an active learning algorithm that sequentially selects the input to excite the system according to an experiment design criterion using the past measured data. The adopted criterion yields a non-convex optimization problem, but we provide an exact convex reformulation allowing to find the global optimizer in a computationally tractable way. Moreover, we give sample complexity bounds on the estimation error due to the stochastic noise. Numerical studies showcase the effectiveness of our algorithm and the benefits of the convex reformulation.

eess.SY↗