Optimization of Functional Materials Design with Optimal Initial Data in Surrogate-Based Active Learning

2025-06-03

AI Summary

In optimizing complex geometric structures of functional materials, the initial dataset size is often ill-suited to multi-scale design spaces, hindering efficient surrogate-model-based optimization. Method: This paper proposes a surrogate-model-driven active learning initialization strategy. It systematically models the quantitative relationship between design-space scale and optimal initial sample size, innovatively employing piecewise linear regression to automatically identify the convergence onset point; “initial data sufficiency” is thereby established as a critical prerequisite for efficiency. The approach integrates a factorized QUBO surrogate model, iterative active sampling, and a quantum-computing-augmented framework. Contribution/Results: Experiments demonstrate accelerated surrogate-model convergence across multi-scale design spaces, reducing computational cost by 30–50%. The strategy delivers the first scalable, reusable initialization protocol for high-throughput materials discovery.

Technology Category

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

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

The optimization of functional materials is important to enhance their properties, but their complex geometries pose great challenges to optimization. Data-driven algorithms efficiently navigate such complex design spaces by learning relationships between material structures and performance metrics to discover high-performance functional materials. Surrogate-based active learning, continually improving its surrogate model by iteratively including high-quality data points, has emerged as a cost-effective data-driven approach. Furthermore, it can be coupled with quantum computing to enhance optimization processes, especially when paired with a special form of surrogate model ($i.e.$, quadratic unconstrained binary optimization), formulated by factorization machine. However, current practices often overlook the variability in design space sizes when determining the initial data size for optimization. In this work, we investigate the optimal initial data sizes required for efficient convergence across various design space sizes. By employing averaged piecewise linear regression, we identify initiation points where convergence begins, highlighting the crucial role of employing adequate initial data in achieving efficient optimization. These results contribute to the efficient optimization of functional materials by ensuring faster convergence and reducing computational costs in surrogate-based active learning.

Problem

• Optimizing functional materials with complex geometries
• Determining optimal initial data sizes for efficient convergence
• Enhancing surrogate-based active learning with quantum computing

Innovation

• Surrogate-based active learning optimizes materials design
• Quantum computing enhances optimization with surrogate models
• Averaged regression determines optimal initial data sizes