I’m a programmer / computer scientist interested in both data efficient machine learning and hardware development.
Currently, I am researching machine learning methods for brain-computer interfaces during my PhD studies. In this field, linear discriminant analysis is a very popular method, which requires the estimation of a covariance matrix. As the latter can become extremely large, I proposed some new approaches to estimate the covariance matrix for a specific type of brain signals (event-related potentials). I also proposed a method to estimate experimental parameters of a BCI during the online experiment to adapt to the current user. See my publications and GitHub profile for details.
In my spare time I like to do stuff with hardware, especially using some sensors to do something (more or less) useful. The idea of this blog was to document especially these embedded projects I’m doing, but so far I have not posted too much in this regard, as you can see from my empty post archive… :)
Other than that I like to bake bread and programmed a software that helps me to easily plan long running recipes, e.g., three-step sourdough bread, so that I don’t have to get up in the middle of the night to tend to the dough. Additional features are scaling of recipes, displaying dough yield of current recipe to judge viscosity of the dough, visualization of the recipe using dot/graphviz. Once I cleaned up the repository and removed recipes I copied from bought recipe books, I plan to post it on my public GitHub.