We are excited to invite you to the next seminar of the 2025 Geology and Geophysics Seminar Series (formerly Earthbyte Seminars), featuring Diange Zhou a PhD candidate from China University of Geosciences. Diange Zhou will be presenting on “Mineral Spatial Distribution Prediction based on Representation Learning” sharing fresh insights and innovative methods in geoscience.
Date: February 5, 2025
Time: 11:00 a.m. – 12:00 p.m. AEDT
Location: Online (Join via zoom)
In-person Viewing: Room 449 (Conference Room), Madsen Building (F09), School of Geosciences
Note: This will be an online talk, but feel free to also join in Conference Room 449
We look forward to seeing you there in person or joining us online!
https://uni-sydney.zoom.us/j/83224746519?from=addon
Mineral Spatial Distribution Prediction based on Representation Learning
Abstract
The mineral spatial distribution prediction seeks to identify potential mineral locations by integrating observational data, thus supporting practical exploration efforts. Previous research has concentrated on modeling the nonlinear relationships between geological, geochemical, geophysical, and remote sensing data in relation to mineralization. The accuracy and low-noise nature of the initial control features are crucial for the effectiveness of these methods. However, the collection of such geoscientific data is both costly and labor-intensive, which limits their flexibility and scalability in real-world applications. To address this issue, this paper introduces the representation learning theory to improve mineral spatial distribution prediction and proposes a novel method, Mineral Representation Learning for Mineral Resource Prediction (MRLMRP). A series of representation learning strategies, including negative sampling, contrastive learning, and node reconstruction, are combined to optimize the prediction of mineral resources. This study offers an innovative approach to mineral resource prediction and provides a methodological reference for various prediction tasks in the geological field.
Graphical Abstract
