Curiosity Driven Exploration to Optimize Structure-Property Learning in Microscopy

2025-04-28

AI Summary

To address low structural–property mapping efficiency and high computational costs in multidimensional output modeling for materials microstructural imaging, this paper proposes a lightweight curiosity-driven active sampling method. Departing from conventional deep-kernel active learning—known for its high computational overhead—the method introduces a novel curiosity-sampling paradigm grounded in uncertainty quantification from surrogate model predictions, prioritizing unexplored, high-uncertainty regions in the structural–property space. By jointly leveraging microscopic images and spectral data, it enables efficient, uncertainty-guided experimental design. In structural-to-property prediction tasks, the method achieves higher mapping accuracy with significantly fewer experiments compared to random sampling, markedly improving sampling efficiency and cross-domain generalizability. This work establishes a new paradigm for interpretable and deployable intelligent materials design.

Technology Category

• Machine Learning: Active Learning
• Search and Optimization: Sampling/Simulation-based Search
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Web Mining and Content Analysis: Bridging structured and unstructured data
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

Rapidly determining structure-property correlations in materials is an important challenge in better understanding fundamental mechanisms and greatly assists in materials design. In microscopy, imaging data provides a direct measurement of the local structure, while spectroscopic measurements provide relevant functional property information. Deep kernel active learning approaches have been utilized to rapidly map local structure to functional properties in microscopy experiments, but are computationally expensive for multi-dimensional and correlated output spaces. Here, we present an alternative lightweight curiosity algorithm which actively samples regions with unexplored structure-property relations, utilizing a deep-learning based surrogate model for error prediction. We show that the algorithm outperforms random sampling for predicting properties from structures, and provides a convenient tool for efficient mapping of structure-property relationships in materials science.

Problem

• Optimizing structure-property learning in microscopy efficiently
• Reducing computational cost in multi-dimensional output spaces
• Enhancing material science via curiosity-driven exploration

Innovation

• Lightweight curiosity algorithm for active sampling
• Deep-learning surrogate model for error prediction
• Efficient mapping of structure-property relationships