FTIR QA/QC with genAI: Application to the Minerals Industry

Speaker

Dr Chris Gonzalez

Abstract

Generative models have gained attention in recent years for their ability to create things like art, music, and realistic impersonations (e.g. faces). This presentation explores how we can leverage that capability in order detect outliers or anomalies within geological datasets. We examine the use of a class of generative models called 'variational autoencoders' to detect when a FTIR spectrum, classed as an unmineralized BIF from rock chips, may have been accidentally mislabelled. This type of work is important because it can be done prior (yes, it lives in the Bayesian world!) to being integrated into larger decision making models, whereby the outliers may invoke wrong predictions. Wrong predictions may result in significant consequences later on in the value chain and/or during the material processing stage. We show how this method offered an initial screening of data in situations where it may not be possible for human intervention to keep up or analyse other spectra coming in rapidly.

Bio

Dr Chris Gonzalez is a Research Fellow in the School of Earth Sciences at the University of Western Australia. He holds many interests, however they all mainly focus on applying computational solutions to complex geological problems. His research has investigated all scales from hand samples to global deep carbon cycles in subduction and intracratonic geological settings. Chris's primary focus now is on developing anomaly and out-of-distribution detection methods for geophysical and geochemical datasets applying state-of-the art computational methods. His role in the CCRF is leading the design and deployment of secure, scalable data storage and analytics platforms tailored to complex geoscientific datasets. 

This seminar will be presented online only