Machine Learning of Two-Dimensional Spectroscopic Data

TL;DR

This paper demonstrates how neural networks can efficiently extract structural and dynamical parameters from two-dimensional electronic spectra, replacing computationally intensive methods and revealing complex disorder effects.

Contribution

It introduces a neural network approach to infer model parameters and generate spectra from 2DES data, integrating exact theoretical models into efficient prediction tools.

Findings

Neural networks can accurately predict model parameters from 2DES data.

Disorder averaging significantly affects polarization-controlled 2DES.

The method enables fast computation of disordered averaged spectra.

Abstract

Two-dimensional electronic spectroscopy has become one of the main experimental tools for analyzing the dynamics of excitonic energy transfer in large molecular complexes. Simplified theoretical models are usually employed to extract model parameters from the experimental spectral data. Here we show that computationally expensive but exact theoretical methods encoded into a neural network can be used to extract model parameters and infer structural information such as dipole orientation from two dimensional electronic spectra (2DES) or reversely, to produce 2DES from model parameters. We propose to use machine learning as a tool to predict unknown parameters in the models underlying recorded spectra and as a way to encode computationally expensive numerical methods into efficient prediction tools. We showcase the use of a trained neural network to efficiently compute disordered averaged spectra and demonstrate that disorder averaging has non-trivial effects for polarization controlled 2DES.