Optimal linear response for expanding circle maps

TL;DR

This paper develops a theoretical and computational framework for identifying optimal linear perturbations of expanding circle maps that maximize the response of observables or spectral properties, with practical Fourier-based methods.

Contribution

It introduces a unique optimal perturbation concept for expanding circle maps and provides explicit Fourier coefficient expressions and computational schemes.

Findings

Existence and uniqueness of optimal perturbations.

Explicit formulas for optimal perturbations in Fourier space.

A Fourier-based computational method demonstrated on examples.

Abstract

We consider the problem of optimal linear response for deterministic expanding maps of the circle. To each infinitesimal perturbation $\dot{T}$ of a circle map $T$ we consider (i) the response of the expectation of an observation function and (ii) the response of isolated spectral points of the transfer operator of $T$. In each case, under mild conditions on the set of feasible perturbations $\dot{T}$ we show there is a unique optimal feasible infinitesimal perturbation $\dot{T}_{\rm optimal}$, maximising the increase of the expectation of the given observation function or maximising the increase of the spectral gap of the transfer operator associated to the system. We derive expressions for the unique maximiser $\dot{T}_{\rm optimal}$ in terms of its Fourier coefficients. We also devise a Fourier-based computational scheme and apply it to illustrate our theory.