Simulation of dielectric axion haloscopes with deep neural networks: a proof-of-principle

TL;DR

This paper demonstrates that deep neural networks can effectively assist in simulating and optimizing dielectric axion haloscopes, potentially reducing computational costs for dark matter axion searches.

Contribution

It introduces the novel application of deep learning techniques to simulate and optimize dielectric haloscopes, addressing computational challenges.

Findings

Deep neural networks can accurately simulate dielectric haloscope responses.

The approach reduces computational resources needed for haloscope design.

Potential for improved efficiency in dark matter axion detection experiments.

Abstract

Dielectric axion haloscopes, such as the MADMAX experiment, are promising concepts for the direct search for dark matter axions. A reliable simulation is a fundamental requirement for the successful realisation of the experiments. Due to the complexity of the simulations, the demands on computing resources can quickly become prohibitive. In this paper, we show for the first time that modern deep learning techniques can be applied to aid the simulation and optimisation of dielectric haloscopes.