Diversity-Aware Batch-Mode Active Learning for Efficient Sampling in Data-Driven Constitutive Modeling
Abstract
The constitutive behavior of materials is modeled through relationships between stress, strain, and possibly additional internal variables. This results in relatively high-dimensional feature spaces for machine learning models rendering the efficient generation of informative datasets essential as brute force methods suffer from the curse of dimensionality. This work introduces a diversity-aware batch-mode query-by-committee active-learning strategy to generate datasets of maximum information content at minimum cost. In contrast to existing methods, this novel method selects multiple informative, non-redundant queries per iteration, enabling concurrent generation of informative datasets and reducing the number of machine-learning retraining cycles. A central component of this method is a cosine-similarity-based metric that complements the uncertainty criterion based on committee variance by promoting within-batch diversity. The query selection is guided by committee variance and a diversity-promoting criterion. The approach is benchmarked for efficient stress-space sampling in data-driven constitutive modeling. In this setting, a committee of support vector classifiers approximates the so-called yield surface, which is a manifold dividing the six-dimensional stress space into an elastic and plastic domain. We demonstrate that the method handles different batch sizes robustly, maintains high within batch diversity, and rapidly reduces committee uncertainty. The resulting machine learning yield surfaces achieve predictive accuracy comparable to sequential active learning, while requiring substantially fewer retraining cycles. This makes the proposed approach an efficient strategy for stress space sampling in data driven constitutive modeling and for reducing time to solution via concurrent data collection in each iteration.
Versions
Related
- Physics-Informed Residual Deep Learning for Constitutive Modeling of Hot Deformation and Dynamic Recrystallization in a Mo-Rich α+β Titanium Alloy
- DNNsolver: Accurate and Efficient Deep Learning Modeling of Polarization-Sensitive Diffractive Neural Networks
- A micromorphic constitutive framework for (non)linear softening modeling
- Energy driven constitutive modeling of thermally damaged red sandstone under multilevel cyclic loading
- A deviatoric-stress closure for constitutive modeling of viscoelastic dynamics
BibTeX
@misc{shoghi2026diversityaware,
title = {{Diversity-Aware Batch-Mode Active Learning for Efficient Sampling in Data-Driven Constitutive Modeling}},
author = {Ronak Shoghi and Lukas Morand and Dirk Helm and Alexander Hartmaier},
howpublished = {arXiv},
year = {2026},
month = may,
doi = {10.48550/arxiv.2605.19993},
eprint = {2605.19993},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2605.19993},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Diversity-Aware Batch-Mode Active Learning for Efficient Sampling in Data-Driven Constitutive Modeling
AU - Ronak Shoghi
AU - Lukas Morand
AU - Dirk Helm
AU - Alexander Hartmaier
PY - 2026
DA - 2026/05/19
PB - arXiv
DO - 10.48550/arxiv.2605.19993
UR - https://arxiv.org/abs/2605.19993
AB - The constitutive behavior of materials is modeled through relationships between stress, strain, and possibly additional internal variables. This results in relatively high-dimensional feature spaces for machine learning models rendering the efficient generation of informative datasets essential as brute force methods suffer from the curse of dimensionality. This work introduces a diversity-aware batch-mode query-by-committee active-learning strategy to generate datasets of maximum information content at minimum cost. In contrast to existing methods, this novel method selects multiple informative, non-redundant queries per iteration, enabling concurrent generation of informative datasets and reducing the number of machine-learning retraining cycles. A central component of this method is a cosine-similarity-based metric that complements the uncertainty criterion based on committee variance by promoting within-batch diversity. The query selection is guided by committee variance and a diversity-promoting criterion. The approach is benchmarked for efficient stress-space sampling in data-driven constitutive modeling. In this setting, a committee of support vector classifiers approximates the so-called yield surface, which is a manifold dividing the six-dimensional stress space into an elastic and plastic domain. We demonstrate that the method handles different batch sizes robustly, maintains high within batch diversity, and rapidly reduces committee uncertainty. The resulting machine learning yield surfaces achieve predictive accuracy comparable to sequential active learning, while requiring substantially fewer retraining cycles. This makes the proposed approach an efficient strategy for stress space sampling in data driven constitutive modeling and for reducing time to solution via concurrent data collection in each iteration.
ER -