Physical regularized Hierarchical Generative Model for Metallic Glass Structural Generation and Energy Prediction

2025-05-15

AI Summary

Metallic glasses suffer from high-dimensional configuration spaces and prohibitive computational costs due to the absence of long-range order. Method: We propose GlassVAE, a physics-guided hierarchical graph variational autoencoder. It introduces a novel dual physical regularization scheme—combining radial distribution function (RDF) loss and energy regression loss—to jointly enforce short-/medium-range structural fidelity and physically plausible energy distributions in the latent space. A rotation-, translation-, and permutation-invariant graph neural network enables compact, invariant representation of disordered atomic structures. Contribution/Results: GlassVAE significantly improves the physical plausibility of generated structures and energy prediction accuracy (MAE < 0.05 eV/atom), outperforming existing models. It is the first end-to-end controllable generative framework achieving simultaneous structural reasonableness and energy accuracy for metallic glasses. The model provides an interpretable, generalizable deep generative framework for amorphous material modeling.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

Disordered materials such as glasses, unlike crystals, lack long range atomic order and have no periodic unit cells, yielding a high dimensional configuration space with widely varying properties. The complexity not only increases computational costs for atomistic simulations but also makes it difficult for generative AI models to deliver accurate property predictions and realistic structure generation. In this work, we introduce GlassVAE, a hierarchical graph variational autoencoder that uses graph representations to learn compact, rotation, translation, and permutation invariant embeddings of atomic configurations. The resulting structured latent space not only enables efficient generation of novel, physically plausible structures but also supports exploration of the glass energy landscape. To enforce structural realism and physical fidelity, we augment GlassVAE with two physics informed regularizers, a radial distribution function (RDF) loss that captures characteristic short and medium range ordering and an energy regression loss that reflects the broad configurational energetics. Both theoretical analysis and experimental results highlight the critical impact of these regularizers. By encoding high dimensional atomistic data into a compact latent vector and decoding it into structures with accurate energy predictions, GlassVAE provides a fast, physics aware path for modeling and designing disordered materials.

Problem

• Generating realistic metallic glass atomic structures with AI
• Predicting energy landscapes of disordered materials efficiently
• Ensuring physical fidelity in generative models via physics-informed regularizers

Innovation

• Hierarchical graph variational autoencoder for atomic embeddings
• Physics informed regularizers ensure structural realism
• Compact latent space enables efficient structure generation