A Critical Examination of Active Learning Workflows in Materials Science

TL;DR

This work critically evaluates active learning workflows in materials science, exposing how data scarcity, bias, and inconsistent design choices limit AL effectiveness. It treats AL as a modular closed-loop framework spanning data generation and materials discovery, and analyzes how surrogate models, sampling strategies, uncertainty quantification (UQ), and evaluation metrics interact to drive performance. The authors discuss both model-based and model-free AL, highlight biases such as AL bias and redundancy, and show that improvements in model accuracy do not always translate to accelerated discovery. They offer guidelines for robust, interpretable AL pipelines, including hierarchical sampling, standardized benchmarks, and multi-fidelity considerations, aiming to make AL more reliable and transferable across diverse materials problems.

Abstract

Active learning (AL) plays a critical role in materials science, enabling applications such as the construction of machine-learning interatomic potentials for atomistic simulations and the operation of self-driving laboratories. Despite its widespread use, the reliability and effectiveness of AL workflows depend on implicit design assumptions that are rarely examined systematically. Here, we critically assess AL workflows deployed in materials science and investigate how key design choices, such as surrogate models, sampling strategies, uncertainty quantification and evaluation metrics, relate to their performance. By identifying common pitfalls and discussing practical mitigation strategies, we provide guidance to practitioners for the efficient design, assessment, and interpretation of AL workflows in materials science.

Figures (3)

• Figure 1: Schematic representation of (a) an AL workflow where both model-based and model-free AL strategies can be employed to acquire samples from pool data and update the seed data by interacting with an oracle, (b) various factors that need to be considered while designing the sampling strategy for AL.
• Figure 2: (a) Active learning reduces redundancy in materials datasets: Performance of XGBoost (XGB) and Random Forests (RF) models on band gap prediction trained on datasets obtained by uncertainty guided active learning, pruning, and random sampling from the OQMD14 dataset. Comparable accuracy is achieved with only 10% of the data, highlighting substantial redundancy in the dataset, Adapted with permission from Ref. li2023exploiting, Copyright 2023 Springer (b) schematic representation of active learning bias induced by sampling of data points do not following i.i.d assumption, (c) information-entropy guided active learning (ETAL) minimizing the large structure-stability bias by improving the coverage of less symmetric crystal systems in the JARVIS classical force-field inspired descriptors dataset (top panel) and improved performance of ML models trained on such active learned dataset compared to random sampled dataset (bottom panel). Adapted with permission from Ref. zhang2023entropy, Copyright 2023 American Institute of Physics.
• Figure :