Session C, April 23, 2024, 12:00 noon to 1:30 p.m. ET
Bridging Realms: Artificial Intelligence in Professional Geoscience
Session Chairs: Andrea Waldie, P.Geo., FGC and Robert Hearst, P.Geo.
A moderated Q & A will follow after the panel presentations.
Presentation 1: Demystifying Artificial Intelligence
Speaker: Bethany Edmunds, PhD, Assistant Dean of Computing Programs in Vancouver & Seattle, Northeastern University - SEE SPEAKER'S BIO
SUMMARY
In this talk, Dr. Edmunds will introduce various terminology and concepts in the fields of Artificial Intelligence and Machine Learning. In addition, she will briefly cover the capabilities and risks of the latest technologies. Attendees of this session will be left with an understanding of what is possible and how to start applying it to the field of geoscience.
Presentation 2: Practitioner's Responsibility on the Use of Machine Learning in Mineral Exploration
Speaker: Yuri Kinakin, P.Geo., Rio Tinto Exploration Canada Inc - SEE SPEAKER'S BIO
SUMMARY
With recent improvements in the accessibility of many advanced machine learning and statistical techniques, they are increasingly being applied to mineral explorations. For example, neural networks are now being used to predict geology at scales ranging from continental to that of a single mineral grain. These can be important and useful additions to a geoscientist's tool kit. However, it is the responsibility of a practitioner to understand uncertainty and potential sources of error with any technique they utilize. Several examples will be presented during this talk.
Presentation 3: Navigating the Perils of Data Validation in AI/ML Geoscience
 |
 |
Speaker: Shawn Hood Ph.D. M.Sc. P.Geo. General Manager, ALS Consulting and Data Analytics SEE SPEAKER'S BIO |
Speaker: McLean Trott Director of Orebody Knowledge at ALS GoldSpot Discoveries SEE SPEAKER'S BIO |
SUMMARY
In the burgeoning field of Artificial Intelligence (AI) and Machine Learning (ML) within geoscience, the integrity of input data and the validation of output play pivotal roles in ensuring the reliability and utility of these advanced technologies. McLean Trott and Shawn Hood delve into the critical challenges of data validation in AI/ML applications, emphasizing the perils of comparing disparate datasets ("apples to oranges") and the importance of rigorous data validation processes ("garbage in equals garbage out"). Through their presentation, they will explore the nuances of validating both input data and AI-generated outputs, underscoring the necessity of meticulous data scrutiny to avoid misleading conclusions and enhance the precision of geoscience predictions. By highlighting practical examples and offering insights into best practices, this talk aims to equip practitioners with the knowledge to responsibly leverage AI/ML technologies in mineral exploration, ensuring that these powerful tools augment rather than undermine the quest for geological understanding.
Register online for this session or for the full Symposium.
