ET-AL: Entropy-Targeted Active Learning for Bias Mitigation in Materials Data

Abstract

Growing materials data and data-driven informatics drastically promote the discovery and design of materials. While there are significant advancements in data-driven models, the quality of data resources is less studied despite its huge impact on model performance. In this work, we focus on data bias arising from uneven coverage of materials families in existing knowledge. Observing different diversities among crystal systems in common materials databases, we propose an information entropy-based metric for measuring this bias. To mitigate the bias, we develop an entropy-targeted active learning (ET-AL) framework, which guides the acquisition of new data to improve the diversity of underrepresented crystal systems. We demonstrate the capability of ET-AL for bias mitigation and the resulting improvement in downstream machine learning models. This approach is broadly applicable to data-driven materials discovery, including autonomous data acquisition and dataset trimming to reduce bias, as well as data-driven informatics in other scientific domains.

arXiv comment35 pages, 13 figures, under review

References in corpus (22)

• Scikit-learn: Machine Learning in Python
• SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python
• Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
• Atomistic Line Graph Neural Network for Improved Materials Property Predictions
• Data-driven materials science: status, challenges and perspectives
• Universal Fragment Descriptors for Predicting Electronic Properties of Inorganic Crystals
• On-the-fly Closed-loop Autonomous Materials Discovery via Bayesian Active Learning
• The Joint Automated Repository for Various Integrated Simulations (JARVIS) for data-driven materials design
• GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration
• Unified Representation of Molecules and Crystals for Machine Learning
• Deep Neural Networks for Accurate Predictions of Garnet Stability
• Why do tree-based models still outperform deep learning on tabular data?
• A critical examination of robustness and generalizability of machine learning prediction of materials properties
• Identifying and Correcting Label Bias in Machine Learning
• Autonomous synthesis of metastable materials
• Quantifying the performance of machine learning models in materials discovery
• Database, Features, and Machine Learning Model to Identify Thermally Driven Metal-Insulator Transition Compounds
• Featureless adaptive optimization accelerates functional electronic materials design
• Data preprocessing to mitigate bias: A maximum entropy based approach
• Screening and understanding Li adsorption on 2-dimensional metallic materials by learning physics
• Efficient Estimation of Material Property Curves and Surfaces via Active Learning
• Characterizing Bias in Classifiers using Generative Models

Cited by in corpus (5)

• Data-Driven Design for Metamaterials and Multiscale Systems: A Review
• On the redundancy in large material datasets: efficient and robust learning with less data
• Probing out-of-distribution generalization in machine learning for materials
• Impact of Data Bias on Machine Learning for Crystal Compound Synthesizability Predictions
• Emerging Microelectronic Materials by Design: Navigating Combinatorial Design Space with Scarce and Dispersed Data