Fabricio Arend Torres

Hi, I’m Fabricio, a Computer Scientist working in the area of Machine Learning, Data Science & Statistics. I’m currently a PhD Student at the University of Basel, where I’m researching ways to model different kinds of ornithological data. The applications range from modeling migration events based on radar data to estimating species distribution models of birds based on citizen-science observations.

On the methodological side my interest lie in Gaussian Processes, Density Estimation (including deep generative models and generative flows), and Implicit Representation Networks. Within the University I’m further involved in mentoring Bachelor’s and Master’s theses and assist in the ‘Scientific Computing’ lecture held at our department. Finally, I regularly review papers for some of the major ML Conferences (NeurIPS, ICML, ICLR).

In more practical terms I have worked quite a bit with the most common Machine Learning and Data Science frameworks, notable mentions being Tensorflow, Pytorch and libraries for scalable Gaussian Process (GPflow, GPytorch). Aside from that, I’m comfortable with the general scientific stack of libraries within python (numpy, scipy, pandas, scikit-learn, matplotlib). As the work within my projects includes extensive preprocessing and exploratory analysis, I have some experience working with spatiotemporal geo-data and the data formats common within that area. While there is a wide range of frameworks and languages I haven’t worked with, I always enjoy delving into new technologies and frameworks.

A more complete list of programming languages, frameworks and tools I’ve worked with is listed below. The level of familiarity is indicated with stars, ranging from basic concepts () to extensive experience (**).

• Programming languages: Python (**), R (), Javascript (*)
• Deep / Machine Learning: Tensorflow1 (*) & 2 (). Pytorch (), scikit-learn
• Gaussian Processes: GPflow(**), GPytorch()
• Exploratory Data Analysis & Preprocessing: Python Scientific Stack (**)
• Data Extraction: pandas(), SQL(), Google Earth Engine()
• Workflow: conda/virtenv/pip(), git(), linux(*), SLURM()