Divergence-based characterization of fundamental limitations of adaptive dynamical systems

TL;DR

This paper develops an information-theoretic framework to understand the fundamental limitations of adaptive control and identification in stochastic dynamical systems, linking system complexity, information acquisition, and residual uncertainty.

Contribution

It introduces a meta-theorem connecting metric entropy, online information, and residual uncertainty, providing new minimax bounds and stabilization time estimates.

Findings

Derived new minimax lower bounds on metric identification error.

Established bounds on the minimum time for stabilization of uncertain systems.

Unified the analysis of control and identification limitations using information theory.

Abstract

Adaptive dynamical systems arise in a multitude of contexts, e.g., optimization, control, communications, signal processing, and machine learning. A precise characterization of their fundamental limitations is therefore of paramount importance. In this paper, we consider the general problem of adaptively controlling and/or identifying a stochastic dynamical system, where our {\em a priori} knowledge allows us to place the system in a subset of a metric space (the uncertainty set). We present an information-theoretic meta-theorem that captures the trade-off between the metric complexity (or richness) of the uncertainty set, the amount of information acquired online in the process of controlling and observing the system, and the residual uncertainty remaining after the observations have been collected. Following the approach of Zames, we quantify {\em a priori} information by the Kolmogorov (metric) entropy of the uncertainty set, while the information acquired online is expressed as a sum of information divergences. The general theory is used to derive new minimax lower bounds on the metric identification error, as well as to give a simple derivation of the minimum time needed to stabilize an uncertain stochastic linear system.