About the talk:
The geological history of the planet is a complex puzzle intertwined in space and time with given sparse data and missing information. Machine learning and related datadriven methods are increasingly becoming prominent in Earth and climate sciences. The fusion of machine learning models with geoscientific data, models and expert knowledge can help in better understanding of the planet and guide in resource mining challenges such as mineral exploration with minimal damage to the environment.

In this seminar, we present innovations with advancements in applying machine learning for creating geoscientific models and processing remote sensing data to unravel the geological and climate history of the planet and to aid in prospecting for mineral resources. We first present a framework that couples machine learning models with plate tectonics that unravel the formation of porphyry copper deposits for the past 80 million years in the western edge of America and highlight effective factors in mineralization. We then present an approach that links climatesensitive sedimentary deposits such as coal, evaporites, and glacial deposits to a global plate model via Bayesian machine learning for the last 150 million years. We model the joint distribution of climatesensitive sediments and annual precipitation through geological time and use the dependency between sediments and precipitation to improve the model's predictive accuracy. We then present a deep learningbased approach for processing remote sensing data and creating lithological maps. Finally, we give an overview of ongoing projects and highlight potential areas for collaboration such as data augmentation and uncertainty quantification with a discussion on the challenges and limitations.

About the speaker:
Dr Rohitash Chandra is a Senior Lecturer in Data Science at the UNSW School of Mathematics and Statistics. Dr Chandra leads a program of research encircling methodologies and applications of artificial intelligence; particularly in areas of Bayesian deep learning, neuroevolution, climate extremes, geoscientific models, and mineral exploration. Dr Chandra has developed novel methods for machine learning inspired by neural systems and learning behaviour that include transfer and multitask learning, with the goal of modular deep learning. His current interest is uncertainty quantification and deep learning with applications to language models, vaccine research, and COVID19. Dr Chandra has attracted multimillion dollar funding with a leading international interdisciplinary team. He is the Data Theme Lead of the Australian Research Council (ARC ITTC) Training Centre for Data Analytics in Minerals and Resources (20202025). More information: https://research.unsw.edu.au/people/d...