Generative adversarial linear discriminant analysis (GALDA) is formulated as a broadly applicable tool for reducing overfitting inherent in representative linear dimension reduction algorithms. Inspired by the successes of generative adversarial neural networks (GANs) for minimizing overfitting artifacts in artificial neural networks, GALDA was built around an independent linear algebra framework distinct from those in GANs. Performance metrics were evaluated for GALDA together with other commonly available supervised and unsupervised methods for dimension reduction in simulated spectra generated an open-source Raman database (Romanian Database of Raman Spectroscopy). In contrast to feature extraction and data reduction approaches for minimizing overfitting, GALDA performs data augmentation by identifying and adversarially excluding the regions in spectral space in which genuine data do not reside. Following these proof-of-concept evaluations, dimension reduction based metrics as well as performance in classification were evaluated for GALDA and other common dimension reduction methods for Raman spectra derived from microscopy measurements of microsphereroids of the blood thinner clopidogrel bisulfate.

Generative adversarial linear discriminant analysis for spectroscopy classification

Fall 2022 abstracts, posters, and presentations will launch in mid-September.

ACS Fall 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.