Hidden Convexity in Active Learning: A Convexified Online Input Design for ARX Systems
By: Nicolas Chatzikiriakos , Bowen Song , Philipp Rank and more
Potential Business Impact:
Finds hidden system rules faster using smart tests.
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.
Similar Papers
Closed-Form Input Design for Identification under Output Feedback with Perturbation Constraints
Systems and Control
Adds safe, small wiggles to make systems learn better.
Learning based convex approximation for constrained parametric optimization
Optimization and Control
Solves hard math problems with smart computer brains.
Online and Offline Space-Filling Input Design for Nonlinear System Identification: A Receding Horizon Control-Based Approach
Systems and Control
Makes computer models learn better from less data.