Publikationen

Projekt-Publikationen

• Johannes Hertrich, Fatima Antarou Ba und Gabriele Steidl. Sparse ANOVA Inspired Mixture Models. arXiv: 2105.14893, 2021.
  [ bib | arXiv ]

• Kai Bergermann und Martin Stoll. Matrix function-based centrality measures for layer-coupled mulitplex networks. ArXiv e-prints, 2021.
  [ bib | arXiv ]

• Dominik Alfke, Miriam Gondos, Lucile Peroche und Martin Stoll. A Study of Graph-Based Approaches for Semi-Supervised Time Series Classification. ArXiv e-prints, 2021.
  [ bib | arXiv ]

• Potts, Daniel und Schmischke, Michael. Interpretable Approximation of High-Dimensional Data. arXiv: 2103.13787 [stat.ML], 2021.
  [ bib | arXiv ]

Abschlussarbeiten

• Jeremias Piljug. Bachelorarbeit: Hoch-Dimensionale ANOVA Approximation in Anwendungen. TU Chemnitz, Fakultät für Mathematik, Professur für Angewandte Funktionalanalyis (Prof. D. Potts, M. Schmischke), 2021.
  [ bib ]

Frühere Arbeiten mit Projektbezug (Vorarbeiten)

• Felix Bartel, Michael Schmischke und Daniel Potts. Grouped Transformations in High-Dimensional Explainable ANOVA Approximation. ArXiv e-prints, 2020.
  [ bib | arXiv ]

• Theresa Wagner. Fast Matrix-Vector Multiplication for the ANOVA Kernel. Chemnitz University of Technology, 2020.
  [ bib | pdf ]

• Potts, Daniel und Schmischke, Michael. Learning multivariate functions with low-dimensional structures using polynomial bases. arXiv: 1912.03195 [math.NA], 2019.
  [ bib | arXiv ]

• Potts, Daniel und Schmischke, Michael. Approximation of high-dimensional periodic functions with Fourier-based methods. SIAM J. Numer. Anal., to appear.
  [ bib | pdf ]

• Michael Schmischke. A Fourier approach to learning sparse additive models. Chemnitz University of Technology, 2019.
  [ bib | pdf ]