Particle Swarm Based Hyper-Parameter Optimization for Machine Learned Interatomic Potentials

TL;DR

This paper introduces a particle swarm optimization method to efficiently tune hyper-parameters in machine learning models for interatomic potential energy surfaces, improving automation and model quality in materials research.

Contribution

It proposes a two-step hyper-parameter optimization strategy using a custom particle swarm optimizer for ML-PES generation, reducing computational cost and enhancing automation.

Findings

Two-step HP optimization improves efficiency

Method successfully applied to diverse systems

Reduces number of ML models needed for tuning

Abstract

Modeling non-empirical and highly flexible interatomic potential energy surfaces (PES) using machine learning (ML) approaches is becoming popular in molecular and materials research. Training an ML-PES is typically performed in two stages: feature extraction and structure-property relationship modeling. The feature extraction stage transforms atomic positions into a symmetry-invariant mathematical representation. This representation can be fine-tuned by adjusting on a set of so-called "hyper-parameters" (HPs). Subsequently, an ML algorithm such as neural networks or Gaussian process regression (GPR) is used to model the structure-PES relationship based on another set of HPs. Choosing optimal values for the two sets of HPs is critical to ensure the high quality of the resulting ML-PES model. In this paper, we explore HP optimization strategies tailored for ML-PES generation using a custom-coded parallel particle swarm optimizer (available freely at https://github.com/suresh0807/PPSO.git). We employ the smooth overlap of atomic positions (SOAP) descriptor in combination with GPR-based Gaussian approximation potentials (GAP) and optimize HPs for four distinct systems: a toy C dimer, amorphous carbon, $\alpha$-Fe, and small organic molecules (QM9 dataset). We propose a two-step optimization strategy in which the HPs related to the feature extraction stage are optimized first, followed by the optimization of the HPs in the training stage. This strategy is computationally more efficient than optimizing all HPs at the same time by means of significantly reducing the number of ML models needed to be trained to obtain the optimal HPs. This approach can be trivially extended to other combinations of descriptor and ML algorithm and brings us another step closer to fully automated ML-PES generation.