review seminar sebastian
Machine learning



Sebastian Scher
PhD student at MISU


Machine Learning in Geosciences – Hype or the Future?


Machine learning is a very trending topic. Modern machine learning techniques are part of our everyday life: they are capable of interpreting images, translating speech, detecting credit card fraud, predicting which shoes you will purchase next, and even managed to master the game of Go. All without explicitly being programmed to do so – the computers learned to do so given a huge amount of data.

Geo-sciences are to a large extent data-driven. Therefore, one might ask whether machine learning techniques do (and will) play a role in geosciences as well? Can we get rid of physical models, and replace them by data-trained algorithms?

I will give a general introduction and a historic overview of machine learning and the closely related field of Artificial Intelligence. Then I will discuss past and present uses of machine learning techniques in the geosciences, limitations, and opportunities for future use.


Time and Place

Tuesday June 19th, 11.15
Rossbysalen C609, Arrhenius Laboratory, 6th floor

(This event has taken place.)