- PhD Student
In my research, I'm interested in developing probabilistic machine learning tools and apply them to climate-related problems.
More specifically, my current research is motivated by the following questions:
(1) What will be the temperature at any specific location in 2050 (or any other mid-long term forecast)?
(2) How can we reason about temperatures and their associated uncertainties when we can barely say what will be the temperatures next week?
By answering the first question, we allow countries and regions to adapt to environmental challenges. For example, if we know what the temperatures will be like for the different regions in Botswana up to 2050, then we can also prepare farmers for the changes that rising temperatures might require in animal breeds and crop varieties.
By answering the second question, we create a common language to reason and compare competing temperature forecasts in order to improve them.
Both answers will be addressed through the probabilistic lens. I view probablistic thinking as a methodical approach to answer questions not only about temperatures, but a general approach to reason about the uncertain world around us. Following Laplace's words "Probability theory is nothing but common sense reduced to calculation".
I am supervised by Dr Damon Wischik and Dr Alex Archibald, and am a member of the AI4ER Centre for Doctoral Training.
