My name is Þórir Már Ingólfsson, but you can just call me Thorir.

I have an M.Sc. in Electrical Engineering and Information Technology from ETH Zürich, where I specialized in Systems, Control and Machine Learning.

As a part of my studies at ETH I partook in a semester project, which revolved around implementing high-precision control and localisation for robotic billiard shots. I therefore implemented a vision-based strategy that allowed a robot arm to obtain the excellent cueing accuracy observed in human players. Here is a video of the performance:

Showing the difference between cue camera localization on a variety of shots.

In my Master Thesis, I compared the performance of Temporal Convolutional Networks and Long Short Term Memory on various time series applications. As well as looking into an embedded implementation of both networks with the main focus being on the trade-off between accuracy and energy usage.

Currently, I’m employed at ETH’s Integrated Systems Laboratory as a Scientific Assistant, working on bringing a new network named EEG-TCNet, which I and my master thesis supervisors designed, to an embedded platform.

I’ll use this website for blogging and talking about things that interest me, in particular things revolving around Data Science.

I hope you enjoy it and find it interesting!