About Me

Welcome!

I have moved on from Microsoft Research and am now a Senior Research Scientist at Nuance Communications, working on speech recognition.

From September 2012 and onwards, I was a post-doctoral researcher with Microsoft Research Cambridge (MSRC), working in the Machine Learning and Perception Group headed by Chris Bishop.

From 2006 to 2012, I was a junior researcher at the Austrian Research Institute for Artificial Intelligence (OFAI), working towards my PhD degree.

During spring 2011, I spent three amazing months as a research intern at Microsoft Research Cambridge, working with Sebastian Nowozin and Carsten Rother. This fruitful collaboration was later extended through contract work for MSRC.

The primary objective of my research is to develop efficient methods for structured learning and prediction. At OFAI, my main field of application was natural language processing. Many natural language processing problems are characterized by rich internal structure, so it is necessary to capture the statistical dependencies between output variables in order to make consistent predictions. As one of our principal means of communication, language is also a fascinating topic in itself, and I maintain a strong interest in the topic.

At MSRC, I also became interested in computer vision. While seemingly disparate, many applications in computer vision are in fact very similar to natural language processing problems from a statistical viewpoint. Again, the challenge is to solve high-dimensional prediction problems while incorporating the statistical dependencies between the response variables.

Graphical models are the common language I use to approach structured prediction problems in either field. While extremely versatile and powerful conceptually, most computations in graphical models are NP-hard. For this reason, approximation algorithms are indispensable. Early algorithms in this field work empirically well, but are not yet completely understood and often have serious drawbacks such as non-convergence. In my PhD thesis, I developed various principled approaches that contribute to overcoming of these limitations.

I hold a bachelor’s degree in Software and Information Engineering (2005), a master’s degree in Sofware Engineering & Internet Computing (2008), and a PhD in Electrical Engineering (2012), all from Vienna University of Technology.

In a previous life, I was an embedded software engineer, mostly working on embedded Linux firmware and kernel development. At the INSO group of Vienna University of Technology and a spin-off thereof, RISE, we developed the initial firmware of the Austrian electronic healthcare card reader (e-Card).