Machine learning, Bayesian inference, and time series

Recent years have seen enormous advances in the application of model-based analysis techniques to multivariate time series. One domain of application is neuroimaging, where data are extremely sparse, high-dimensional, and noisy. Another domain is global web traffic, which has remarkably similar structural properties. The design of predictive models for such datasets poses a number of compelling challenges for statistical inference and machine learning.

At present, I work as a Postdoctoral Research Fellow / Quantitative Analyst at Google. In addition, I am affiliated with the Department of Information Technology and Electrical Engineering at the Swiss Federal Institute of Technology / ETH Zurich (Translational Neuromodeling Unit).

Recent publications

Show all publications >>