English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The widespread application of multimodal machine learning models like GPT-4has revolutionized various research fields including computer vision andnatural language processing. However, its implementation in materialsinformatics remains underexplored, despite the presence of materials dataacross diverse modalities, such as composition and structure. The effectivenessof machine learning models trained on large calculated datasets depends on theaccuracy of calculations, while experimental datasets often have limited dataavailability and incomplete information. This paper introduces a novel approachto multimodal machine learning in materials science via composition-structurebimodal learning. The proposed COmposition-Structure Bimodal Network (COSNet)is designed to enhance learning and predictions of experimentally measuredmaterials properties that have incomplete structure information. Bimodallearning significantly reduces prediction errors across distinct materialsproperties including Li conductivity in solid electrolyte, band gap, refractiveindex, dielectric constant, energy, and magnetic moment, surpassingcomposition-only learning methods. Furthermore, we identified that dataaugmentation based on modal availability plays a pivotal role in the success ofbimodal learning.

Multimodal machine learning for materials science: composition-structure bimodal learning for experimentally measured properties