Philipp Thomann
Software Projects
- Bindings to liquidSVM for
R (CRAN),
Python (PyPI),
MATLAB,
Java,
Apache Spark
- plotVR: Walk Through your Data in Cardboard VR (Android)
- NABU: E-learning tool to drill vocabulary (e.g.\ Accadian, Arabic, Chinese, Hindi, Middle Egyptian, Persian, Swahili), many even with sound spoken by native speakers or by Text-to-Speech systems. Also it is easy for teachers to upload new vocabulary.
Research interests
Machine learning, clustering, SVMs, probability theory, statistical mechanics,
natural language processing, data science, simulation of probabilistic models, software development.
Dissertation
- A Generalization of the Sherrington-Kirkpatrick Model, Zurich University, 2012, pdf.
Publications
- Steinwart, I. and Thomann, P. (2017).
liquidSVM: A Fast and Versatile SVM Package.
(arXiv)
- Thomann, P. and Blaschzyk, I. and Meister, M. and Steinwart, I. (2017).
Spatial Decompositions for Large Scale SVMs.
(appearing in AISTATS 2017, arXiv)
- Steinwart, I. and Thomann, P. and Schmid, N. (2016).
Learning with Hierarchical Gaussian Kernels.
(arXiv)
- Thomann, P. and Steinwart, I. and Schmid, N. (2015).
Towards an Axiomatic Approach to Hierarchical Clusterig of Measures.
Journal of Machine Learning Research, Volume 16(Sep), pp 1949--2002
(article, arXiv)
- Thomann, J. and Thomann, P. and Toulany G. and Orthmann E. (2013).
Vokabeltrainer, Vokabeltrainer zum Lehrbuch der persischen Sprache.
Buske (Hamburg).
- Avena, L. and Thomann, P. Continuity and Anomalous Fluctuations in Random Walks in Dynamic Random Environments: Numerics, Phase Diagrams and Conjectures, Journal of Statistical Physics 147 (2012), no. 6, 1041-1067. (article, arXiv)