Bayesian Joint Inversions for the Exploration and Characterization of Geothermal Energy Targets
Event details
| Date | 03.10.2013 |
| Hour | 16:15 › 17:15 |
| Speaker |
Dr. Edwin Bonilla, Machine Learning group at NICTA, Australia Bio: I am a senior researcher at NICTA and an adjunct research fellow at the Australian National University. I work on machine learning, developing probabilistic models and efficient inference methods for structured prediction problems. A non-exhaustive list of my research interests is: Transfer learning Graphical models Bayesian statistics Gaussian processes learning from structured data |
| Location | |
| Category | Conferences - Seminars |
We propose a machine learning approach to geophysical inversion problems for the exploration and characterization of geothermal energy targets. Our approach is based on Bayesian methods, and provides a full distribution over the predicted geophysical properties whilst enabling the incorporation of data from different modalities. We assess our method qualitatively and quantitatively using simulated experiments and a real dataset from South Australia with the goal of characterizing rock properties. The significance of our probabilistic inversion extends to general exploration problems.
Links
Practical information
- Informed public
- Free
Organizer
- Prof. Matthias Seeger
Contact
- Prof. Matthias Seeger