Impact of Data Bias on Machine Learning for Crystal Compound Synthesizability Predictions

Abstract

Machine learning models are susceptible to being misled by biases in training data that emphasize incidental correlations over the intended learning task. In this study, we demonstrate the impact of data bias on the performance of a machine learning model designed to predict the synthesizability likelihood of crystal compounds. The model performs a binary classification on labeled crystal samples. Despite using the same architecture for the machine learning model, we showcase how the model's learning and prediction behavior differs once trained on distinct data. We use two data sets for illustration: a mixed-source data set that integrates experimental and computational crystal samples and a single-source data set consisting of data exclusively from one computational database. We present simple procedures to detect data bias and to evaluate its effect on the model's performance and generalization. This study reveals how inconsistent, unbalanced data can propagate bias, undermining real-world applicability even for advanced machine learning techniques.

8 pages, 4 figures, 3 appendices

References in corpus (12)

• Atomistic Line Graph Neural Network for Improved Materials Property Predictions
• Opportunities and Challenges for Machine Learning in Materials Science
• A critical examination of robustness and generalizability of machine learning prediction of materials properties
• Machine-learning rationalization and prediction of solid-state synthesis conditions
• Assessing Thermodynamic Selectivity of Solid-State Reactions for the Predictive Synthesis of Inorganic Materials
• ET-AL: Entropy-Targeted Active Learning for Bias Mitigation in Materials Data
• Screening and understanding Li adsorption on 2-dimensional metallic materials by learning physics
• Robust model benchmarking and bias-imbalance in data-driven materials science: a case study on MODNet
• Efficient Estimation of Material Property Curves and Surfaces via Active Learning
• Polymorphism in elemental silicon: Probabilistic interpretation of the realizability of metastable structures
• Crystal Structure Generation with Autoregressive Large Language Modeling
• Inorganic Synthesis Reaction Condition Prediction with Generative Machine Learning