Data-Driven Forecasting of Non-Equilibrium Solid-State Dynamics

Abstract

We present a data-driven approach to efficiently approximate nonlinear transient dynamics in solid-state systems. Our proposed machine-learning model combines a dimensionality reduction stage with a nonlinear vector autoregression scheme. We report an outstanding time-series forecasting performance combined with an easy to deploy model and an inexpensive training routine. Our results are of great relevance as they have the potential to massively accelerate multi-physics simulation software and thereby guide to future development of solid-state based technologies.

The simulation code and the regression code is available on GitHub under MIT license (https://github.com/stmeinecke/derrom)

References in corpus (7)

• k.p theory for two-dimensional transition metal dichalcogenide semiconductors
• Next Generation Reservoir Computing
• Intrinsic Transport Properties of Electrons and Holes in Monolayer Transition Metal Dichalcogenides
• Deep Neural Networks for Nonlinear Model Order Reduction of Unsteady Flows
• Hybrid quantum-classical modeling of quantum dot devices
• Strong coupling regime and hybrid quasinormal modes of a single plasmonic resonator coupled to a TMDC monolayer
• Feedback induced locking in semiconductor lasers with strong amplitude-phase coupling