Prospects of reinforcement learning for simultaneous damping of many mechanical modes

Christian Sommer, Muhammad Asjad, Claudiu Genes

Published 2019-05-29Version 1

We apply adaptive feedback for the refrigeration of a mechanical resonator, i.e. with the aim of simultaneously cooling the classical thermal motion of more than one vibrational degree of freedom. The feedback is obtained from a neural network trained via a reinforcement learning strategy to choose the correct sequence of actions from a finite set in order to reduce the total energy of all modes of vibration. The actions are realized either as optical modulations of spring constants or as radiation pressure induced momentum kicks. As a proof of principle we numerically show simultaneous cooling of four independent modes with an overall strong reduction of the total system temperature.