Contraction-Based Methods for Stable Identification and Robust Machine Learning: a Tutorial

TL;DR

This tutorial introduces contraction-based methods for stable and robust machine learning, focusing on system identification and applications like image recognition, emphasizing stability and behavioral guarantees.

Contribution

It adapts contraction analysis and robust control techniques to large-scale machine learning models with stability and robustness guarantees.

Findings

Methods enable stable system identification.

Applications include robust image recognition.

Techniques provide behavioral guarantees.

Abstract

This tutorial paper provides an introduction to recently developed tools for machine learning, especially learning dynamical systems (system identification), with stability and robustness constraints. The main ideas are drawn from contraction analysis and robust control, but adapted to problems in which large-scale models can be learnt with behavioural guarantees. We illustrate the methods with applications in robust image recognition and system identification.