Machine learning researcherIBM Research Australia
Dr Stefan Maetschke received his masters from the University of applied Computer Science in Wuerzburg, Germany and his PhD in Computer Science from the University of Queensland, Australia. He is the holder of several patents and has published over 20 papers in the field of Machine Learning and Computational Biology.
His early career (1990-2001), as a Software Engineer in R&D at Siemens Germany, was focused on non-linear optimization, adaptive quality control and algorithms to improve the trajectorial accuracy of robots.
Following that, he worked for a small company in Germany and was implementing software for speech processing and data mining before joining the PhD programm at the University of Queensland in 2004. During his PhD, Stefan, designed machine learning models to accurately predict the subcellular localization of transmembrane proteins. He completed his PhD in 2007, with the Dean’s commendation for an outstanding thesis.
Until 2013 he was employed as a Research Officer at the Institute for Molecular Bioscience (IMB), Queensland and as a Associate Lecturer and Course Coordinator at the School of Information Technology and Electrical Engineering (ITEE) in Brisbane. The main focus of his research at this time was on computational methods to infer and analyze protein-protein interaction and gene-regulatory networks. He furthermore designed the courses "Advanced Software Engineering" and "Programming in the Large", which he lectured at ITEE and conducted an annual programming course "Introduction into Python for Biologists" at IMB.
In 2013 he was hired by Google, USA to improve the infrastructure and machine learning models for privacy protection of Google Street View imagery; namely the automatic blurring of faces and license plates.
He joined IBM Research Australia in 2016 to develop novel machine learning algorithms for medical image analysis with a focus on anomaly detection.