10 papers
A graph-informed regret metric for optimal distributed control
Daniele Martinelli, Andrea Martin, Giancarlo Ferrari-Trecate +1
We consider the optimal control of large-scale systems using distributed controllers whose network topology mirrors the coupling graph between subsystems. In this work, we introduc…
Learning to optimize with guarantees: a complete characterization of linearly convergent algorithms
Andrea Martin, Ian R. Manchester, Luca Furieri
The design of many classical optimization algorithms is driven by the certification of linear convergence rates over classes of optimization problems. In this paper, we consider th…
Sinkhorn Ambiguity Sets for Distributionally Robust Control: Convexity, Weak Compactness, and Tractability
Riccardo Cescon, Andrea Martin, Giancarlo Ferrari-Trecate
Classical stochastic control assumes perfect knowledge of the uncertainty affecting the plant. In practice, however, such information is often incomplete. To address this limitatio…
On the Global Optimality of Linear Policies for Sinkhorn Distributionally Robust Linear Quadratic Control
Riccardo Cescon, Andrea Martin, Giancarlo Ferrari-Trecate
The Linear Quadratic Gaussian (LQG) regulator is a cornerstone of optimal control theory, yet its performance can degrade significantly when the noise distributions deviate from th…
Learning to accelerate Krasnosel'skii-Mann fixed-point iterations with guarantees
Andrea Martin, Giuseppe Belgioioso
We introduce a principled learning to optimize (L2O) framework for solving fixed-point problems involving general nonexpansive mappings. Our idea is to deliberately inject summable…
MAD: A Magnitude And Direction Policy Parametrization for Stability Constrained Reinforcement Learning
Luca Furieri, Sucheth Shenoy, Danilo Saccani +2
We introduce magnitude and direction (MAD) policies, a policy parameterization for reinforcement learning (RL) that preserves Lp closed-loop stability for nonlinear dynamical syste…