P
Philipp Thomann
Researcher at University of Stuttgart
Publications - 9
Citations - 111
Philipp Thomann is an academic researcher from University of Stuttgart. The author has contributed to research in topics: Support vector machine & Cluster analysis. The author has an hindex of 6, co-authored 9 publications receiving 102 citations. Previous affiliations of Philipp Thomann include University of Zurich.
Papers
More filters
Posted Content
liquidSVM: A Fast and Versatile SVM package
Ingo Steinwart,Philipp Thomann +1 more
TL;DR: This work presents a brief description of the LiquidSVM package, a package written in C++ that provides SVM-type solvers for various classification and regression tasks, and reports experimental comparisons to other SVM packages.
Posted Content
Learning with Hierarchical Gaussian Kernels.
TL;DR: It is shown that Gaussian kernels are universal and that SVMs using these kernels are universally consistent, and a parameter optimization method for the kernel parameters is described that is empirically compared to SVMs, random forests, a multiple kernel learning approach, and to some deep neural networks.
Journal ArticleDOI
Continuity and Anomalous Fluctuations in Random Walks in Dynamic Random Environments: Numerics, Phase Diagrams and Conjectures
Luca Avena,Philipp Thomann +1 more
TL;DR: In this article, the scaling limits of one dimensional continuous-time random walks in two dynamic random environments with fast (independent spin-flips) and slow (simple symmetric exclusion) decay of space-time correlations are investigated.
Journal ArticleDOI
Continuity and anomalous fluctuations in random walks in dynamic random environments: numerics, phase diagrams and conjectures
Luca Avena,Philipp Thomann +1 more
TL;DR: In this paper, the scaling limits of one dimensional continuous-time random walks in two dynamic random environments with fast (independent spin-flips) and slow (simple symmetric exclusion) decay of space-time correlations are investigated.
Proceedings Article
Spatial Decompositions for Large Scale SVMs
TL;DR: In this paper, a decomposition strategy that learns on small, spatially defined data chunks is proposed. And the resulting rates match those known for SVMs solving the complete optimization problem with Gaussian kernels.