NeuralMAG: Fast and Generalizable Micromagnetic Simulation with Deep Neural Nets

TL;DR

NeuralMAG introduces a deep learning-based micromagnetic simulation method that significantly accelerates computations using a U-shaped neural network, enabling fast, large-scale, and generalizable simulations based on the LLG equation.

Contribution

The paper presents NeuralMAG, a novel neural network architecture that accelerates micromagnetic simulations by focusing on core computations, improving speed and scalability over traditional methods.

Findings

Achieves O(N) time complexity for demagnetizing field computation.

Validated on multiple tasks with different sample sizes, shapes, and materials.

Outperforms traditional FFT-based methods in speed for large-scale simulations.

Abstract

Micromagnetics has made significant strides, particularly due to its wide-ranging applications in magnetic storage design. Numerical simulation is a cornerstone of micromagnetics research, relying on first-principle rules to compute the dynamic evolution of micromagnetic systems based on the renowned LLG equation, named after Landau, Lifshitz, and Gilbert. However, simulations are often hindered by their slow speed. Although Fast-Fourier transformation (FFT) calculations reduce the computational complexity to O(NlogN), it remains impractical for large-scale simulations. In this paper, we introduce NeuralMAG, a deep learning approach to micromagnetic simulation. Our approach follows the LLG iterative framework but accelerates demagnetizing field computation through the employment of a U-shaped neural network (Unet). The Unet architecture comprises an encoder that extracts aggregated spins at various scales and learns the local interaction at each scale, followed by a decoder that accumulates the local interactions at different scales to approximate the global convolution. This divide-and-accumulate scheme achieves a time complexity of O(N), significantly enhancing the speed and feasibility of large-scale simulations. Unlike existing neural methods, NeuralMAG concentrates on the core computation rather than an end-to-end approximation for a specific task, making it inherently generalizable. To validate the new approach, we trained a single model and evaluated it on two micromagnetics tasks with various sample sizes, shapes, and material settings.