Microstructure representation learning using Siamese networks. (December 2020)

• Record Type: Journal Article
• Authors: Sardeshmukh, Avadhut; Reddy, Sreedhar; Gautham, B.P.; Bhattacharyya, Pushpak
• Is Part Of: MRS communications. Volume 10:Number 4(2020)
• Page Start: 613
• Page End: 619
• Publication Date: 2020-12
• Subjects: Materials -- Periodicals; 620.11
• DOI: 10.1557/mrc.2020.55
• Languages: English
• ISSNs: 2159-6859

Abstract

Abstract : Abstract: Obtaining a good statistical representation of material microstructures is crucial for establishing robust process–structure–property linkages and machine learning techniques can bridge this gap. One major difficulty in leveraging recent advances in deep learning for this purpose is the scarcity of good quality data with enough metadata. In machine learning, similarity metric learning using Siamese networks has been used to deal with sparse data. Inspired by this, the authors propose a Siamese architecture to learn microstructure representations. The authors show that analysis tasks such as the classification of microstructures can be done more efficiently in the learned representation space.