Self-Supervised Learning for Glass Composition Screening

TL;DR

This paper introduces a self-supervised learning framework with a novel neural network architecture for accurately screening glass compositions based on their transition temperatures, overcoming data noise issues and improving generalization in material discovery.

Contribution

It presents DeepGlassNet, a new self-supervised model architecture, and a data augmentation strategy for robust glass composition screening within transition temperature ranges.

Findings

DeepGlassNet outperforms traditional screening methods in accuracy.

The model demonstrates strong adaptability to other composition-related tasks.

The approach effectively handles noisy data in glass material datasets.

Abstract

Glass composition screening is essential for advancing new glass materials, yet the inherent complexity of multicomponent systems presents significant challenges. Current supervised learning methods for this task rely heavily on large amounts of high-quality data and are prone to overfitting on noisy samples, which limits their generalization ability. In this work, we propose a novel self-supervised learning framework designed specifically for screening glass compositions within pre-defined glass transition temperature (Tg) ranges. We reformulate the screening task as a classification problem, aiming to predict whether the glass transition temperature of a given composition falls within a target interval. To improve the model's robustness to noise, we introduce an innovative data augmentation strategy grounded in asymptotic theory. Additionally, we present DeepGlassNet, a dedicated network architecture developed to capture and analyze the complex interactions among constituent elements in glass compositions. Experimental results demonstrate that DeepGlassNet achieves superior screening accuracy compared to traditional methods and exhibits strong adaptability to other composition-related screening tasks. This study not only provides an efficient methodology for designing multicomponent glasses but also establishes a foundation for applying self-supervised learning in material discovery. Code and data are available at: https://github.com/liubin06/DeepGlassNet