Machine learning on the electron-boson mechanism in superconductors

TL;DR

This paper demonstrates that machine learning, specifically neural networks and autoencoders, can infer the electron-boson pairing mechanism in superconductors from limited experimental data, effectively capturing complex relationships.

Contribution

It introduces a machine learning framework combining neural networks and autoencoders to infer pairing mechanisms from superconducting gap and spectral functions, handling complex data.

Findings

Neural networks accurately predict spectral functions for simple cases.

Autoencoders effectively reduce complexity for complex spectral functions.

Method applicable to general function-to-function mappings in physics and beyond.

Abstract

To unravel pairing mechanism of a superconductor from limited, indirect experimental data is always a difficult task. It is common but sometimes dubious to explain by a theoretical model with some tuning parameters. In this work, we propose that the machine learning might infer pairing mechanism from observables like superconducting gap functions. For superconductivity within the Migdal-Eliashberg theory, we perform supervised learning between superconducting gap functions and electron-boson spectral functions. For simple spectral functions, the neural network can easily capture the correspondence and predict perfectly. For complex spectral functions, an autoencoder is utilized to reduce the complexity of the spectral functions to be compatible to that of the gap functions. After this complexity-reduction process, relevant information of the spectral function is extracted and good performance restores. Our proposed method can extract relevant information from data and can be applied to general function-to-function mappings with asymmetric complexities either in physics or other fields.