Anastasios Tsiamis
YOU?
Author Swipe
View article: Policy Gradient Bounds in Multitask LQR
Policy Gradient Bounds in Multitask LQR Open
We analyze the performance of policy gradient in multitask linear quadratic regulation (LQR), where the system and cost parameters differ across tasks. The main goal of multitask LQR is to find a controller with satisfactory performance on…
View article: Layered Multirate Control of Constrained Linear Systems
Layered Multirate Control of Constrained Linear Systems Open
Layered control architectures have been a standard paradigm for efficiently managing complex constrained systems. A typical architecture consists of: i) a higher layer, where a low-frequency planner controls a simple model of the system, a…
View article: Distributionally Robust Optimization over Wasserstein Balls with i.i.d. Structure
Distributionally Robust Optimization over Wasserstein Balls with i.i.d. Structure Open
We consider distributionally robust optimization problems where the uncertainty is modeled via a structured Wasserstein ambiguity set. Specifically, the ambiguity is restricted to product measures $P^{\otimes N}$, where $P$ lies within a W…
View article: Semismooth Newton Methods for Risk-Averse Markov Decision Processes
Semismooth Newton Methods for Risk-Averse Markov Decision Processes Open
Inspired by semismooth Newton methods, we propose a general framework for designing solution methods with convergence guarantees for risk-averse Markov decision processes. Our approach accommodates a wide variety of risk measures by levera…
View article: Operator Splitting for Convex Constrained Markov Decision Processes
Operator Splitting for Convex Constrained Markov Decision Processes Open
We consider finite Markov decision processes (MDPs) with convex constraints and known dynamics. In principle, this problem is amenable to off-the-shelf convex optimization solvers, but typically this approach suffers from poor scalability.…
View article: Online Residual Learning from Offline Experts for Pedestrian Tracking
Online Residual Learning from Offline Experts for Pedestrian Tracking Open
In this paper, we consider the problem of predicting unknown targets from data. We propose Online Residual Learning (ORL), a method that combines online adaptation with offline-trained predictions. At a lower level, we employ multiple offl…
View article: Data-Driven Distributionally Robust System Level Synthesis
Data-Driven Distributionally Robust System Level Synthesis Open
We present a novel approach for the control of uncertain, linear time-invariant systems, which are perturbed by potentially unbounded, additive disturbances. We propose a \emph{doubly robust} data-driven state-feedback controller to ensure…
View article: On the Regret of Recursive Methods for Discrete-Time Adaptive Control with Matched Uncertainty
On the Regret of Recursive Methods for Discrete-Time Adaptive Control with Matched Uncertainty Open
Continuous-time adaptive controllers for systems with a matched uncertainty often comprise an online parameter estimator and a corresponding parameterized controller to cancel the uncertainty. However, such methods are often impossible to …
View article: Finite Sample Frequency Domain Identification
Finite Sample Frequency Domain Identification Open
We study non-parametric frequency-domain system identification from a finite-sample perspective. We assume an open loop scenario where the excitation input is periodic and consider the Empirical Transfer Function Estimate (ETFE), where the…
View article: Predictive Linear Online Tracking for Unknown Targets
Predictive Linear Online Tracking for Unknown Targets Open
In this paper, we study the problem of online tracking in linear control systems, where the objective is to follow a moving target. Unlike classical tracking control, the target is unknown, non-stationary, and its state is revealed sequent…
View article: The Fundamental Limitations of Learning Linear-Quadratic Regulators
The Fundamental Limitations of Learning Linear-Quadratic Regulators Open
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 of a stabilizing controller and an energy bounded e…
View article: A Tutorial on the Non-Asymptotic Theory of System Identification
A Tutorial on the Non-Asymptotic Theory of System Identification Open
This tutorial serves as an introduction to recently developed non-asymptotic methods in the theory of -- mainly linear -- system identification. We emphasize tools we deem particularly useful for a range of problems in this domain, such as…
View article: The Fundamental Limitations of Learning Linear-Quadratic Regulators
The Fundamental Limitations of Learning Linear-Quadratic Regulators Open
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 of a stabilizing controller and an energy bounded e…
View article: Online Linear Quadratic Tracking with Regret Guarantees
Online Linear Quadratic Tracking with Regret Guarantees Open
Online learning algorithms for dynamical systems provide finite time guarantees for control in the presence of sequentially revealed cost functions. We pose the classical linear quadratic tracking problem in the framework of online optimiz…
View article: Suboptimality analysis of receding horizon quadratic control with unknown linear systems and its applications in learning-based control
Suboptimality analysis of receding horizon quadratic control with unknown linear systems and its applications in learning-based control Open
This work analyzes how the trade-off between the modeling error, the terminal value function error, and the prediction horizon affects the performance of a nominal receding-horizon linear quadratic (LQ) controller. By developing a novel pe…
View article: Implications of Regret on Stability of Linear Dynamical Systems
Implications of Regret on Stability of Linear Dynamical Systems Open
The setting of an agent making decisions under uncertainty and under dynamic constraints is common for the fields of optimal control, reinforcement learning, and recently also for online learning. In the online learning setting, the qualit…
View article: How are policy gradient methods affected by the limits of control?
How are policy gradient methods affected by the limits of control? Open
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 inevitably lead to noisy gradient estimates. We also give an ex…
View article: Implications of Regret on Stability of Linear Dynamical Systems
Implications of Regret on Stability of Linear Dynamical Systems Open
The setting of an agent making decisions under uncertainty and under dynamic constraints is common for the fields of optimal control, reinforcement learning, and recently also for online learning. In the online learning setting, the qualit…
View article: Statistical Learning Theory for Control: A Finite Sample Perspective
Statistical Learning Theory for Control: A Finite Sample Perspective Open
This tutorial survey provides an overview of recent non-asymptotic advances in statistical learning theory as relevant to control and system identification. While there has been substantial progress across all areas of control, the theory …
View article: Secure state estimation over Markov wireless communication channels (extended version)
Secure state estimation over Markov wireless communication channels (extended version) Open
This note studies state estimation in wireless networked control systems with secrecy against eavesdropping. Specifically, a sensor transmits a system state information to the estimator over a legitimate user link, and an eavesdropper over…
View article: How are policy gradient methods affected by the limits of control?
How are policy gradient methods affected by the limits of control? Open
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 inevitably lead to noisy gradient estimates. We also give an ex…
View article: Learning to Control Linear Systems can be Hard
Learning to Control Linear Systems can be Hard Open
In this paper, we study the statistical difficulty of learning to control linear systems. We focus on two standard benchmarks, the sample complexity of stabilization, and the regret of the online learning of the Linear Quadratic Regulator …
View article: State-Output Risk-Constrained Quadratic Control of Partially Observed Linear Systems
State-Output Risk-Constrained Quadratic Control of Partially Observed Linear Systems Open
We propose a methodology for performing risk-averse quadratic regulation of partially observed Linear Time-Invariant (LTI) systems disturbed by process and output noise. To compensate against the induced variability due to both types of no…
View article: Adaptive Stochastic MPC under Unknown Noise Distribution
Adaptive Stochastic MPC under Unknown Noise Distribution Open
In this paper, we address the stochastic MPC (SMPC) problem for linear systems, subject to chance state constraints and hard input constraints, under unknown noise distribution. First, we reformulate the chance state constraints as determi…
View article: Linear Systems can be Hard to Learn
Linear Systems can be Hard to Learn Open
In this paper, we investigate when system identification is statistically easy or hard, in the finite sample regime. Statistically easy to learn linear system classes have sample complexity that is polynomial with the system dimension. Mos…
View article: Encrypted Distributed Lasso for Sparse Data Predictive Control
Encrypted Distributed Lasso for Sparse Data Predictive Control Open
The least squares problem with L1-regularized regressors, called Lasso, is a widely used approach in optimization problems where sparsity of the regressors is desired. This formulation is fundamental for many applications in signal process…
View article: Risk-Constrained Linear-Quadratic Regulators
Risk-Constrained Linear-Quadratic Regulators Open
We propose a new risk-constrained reformulation of the standard Linear Quadratic Regulator (LQR) problem. Our framework is motivated by the fact that the classical (risk-neutral) LQR controller, although optimal in expectation, might be in…
View article: Sparse Approximate Solutions to Max-Plus Equations with Application to Multivariate Convex Regression
Sparse Approximate Solutions to Max-Plus Equations with Application to Multivariate Convex Regression Open
In this work, we study the problem of finding approximate, with minimum support set, solutions to matrix max-plus equations, which we call sparse approximate solutions. We show how one can obtain such solutions efficiently and in polynomia…
View article: Data-driven control on encrypted data
Data-driven control on encrypted data Open
We provide an efficient and private solution to the problem of encryption-aware data-driven control. We investigate a Control as a Service scenario, where a client employs a specialized outsourced control solution from a service provider. …