Active Learning for Conditional Inverse Design with Crystal Generation and Foundation Atomic Models

2025-02-24

AI Summary

Current inverse materials design suffers from insufficient synergy between crystal generation and property prediction models. Method: We propose an active learning framework that integrates a crystal generation model with a foundational atomic-scale model. Our approach introduces a novel “generate–evaluate–feedback” iterative paradigm, incorporating the Con-CDVAE crystal generator, the MACE-MP-0 atomic-scale potential model, and an uncertainty-aware active learning strategy to enable property-driven fine-tuning and cross-task generalization. Contribution/Results: Experiments demonstrate significant improvements in targeted generation of high-bulk-modulus crystals: the framework achieves high accuracy and converges efficiently to target structures within a single iteration. It balances computational efficiency, predictive reliability, and model scalability—establishing a new paradigm for inverse materials design under data-scarce conditions.

Technology Category

• Machine Learning: Active Learning
• Search and Optimization: Sampling/Simulation-based Search
• Natural Language Processing: Generation

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

Artificial intelligence (AI) is transforming materials science, enabling both theoretical advancements and accelerated materials discovery. Recent progress in crystal generation models, which design crystal structures for targeted properties, and foundation atomic models (FAMs), which capture interatomic interactions across the periodic table, has significantly improved inverse materials design. However, an efficient integration of these two approaches remains an open challenge. Here, we present an active learning framework that combines crystal generation models and foundation atomic models to enhance the accuracy and efficiency of inverse design. As a case study, we employ Con-CDVAE to generate candidate crystal structures and MACE-MP-0 FAM as one of the high-throughput screeners for bulk modulus evaluation. Through iterative active learning, we demonstrate that Con-CDVAE progressively improves its accuracy in generating crystals with target properties, highlighting the effectiveness of a property-driven fine-tuning process. Our framework is general to accommodate different crystal generation and foundation atomic models, and establishes a scalable approach for AI-driven materials discovery. By bridging generative modeling with atomic-scale simulations, this work paves the way for more accurate and efficient inverse materials design.

Problem

• Integration of crystal generation and atomic models
• Enhancing inverse materials design accuracy
• Scalable AI-driven materials discovery approach

Innovation

• Active learning framework integration
• Crystal generation with Con-CDVAE
• Foundation atomic models for screening