Schnelleinstieg Reader

Home|Suche|Sitemap|Webmail

Startseite FSU





franknu

Frank Nussbaum

PhD Student

Lehrstuhl für Theoretische Informatik II
Fakultät für Mathematik und Informatik
Friedrich-Schiller-Universität Jena

Ernst-Abbe-Platz 2
D-07743 Jena
Germany

Office: 3335
Tel.: +49 3641 9 46328
Fax: +49 3641 9 46322
E-mail: frank dot nussbaum at uni-jena dot de

Research Interests

My primary interest lies in statistical models for machine learning. My goal is to obtain a deeper understanding of currently available methods. I intend to create quality models that are generically applicable and easy to use. My research interests include in particular:

  • multivariate probability distributions (latent variable graphical models, SPNs)
  • convex optimization solvers (ADMM, proximal point methods)
  • causality

Publications

Conference Proceedings
  • F. Nussbaum and J. Giesen. Disentangling Direct and Indirect Interactions in Polytomous Item Response Theory Models. To appear in: Proceedings of the 29th International Joint Conference on Artificial Intelligence (IJCAI), (2020)    Acceptance Rate: 12.6%
  • J. Giesen, F. Nussbaum and C. Schneider. Efficient Regularization Parameter Selection for Latent Variable Graphical Models via Bi-Level Optimization. Proceedings of the 28th International Joint Conference on Artificial Intelligence (IJCAI), (2019)    Acceptance Rate: 17.9%
  • F. Nussbaum and J. Giesen. Ising Models with Latent Conditional Gaussian Variables. Proceedings of the 30th International Conference on Algorithmic Learning Theory (ALT), (2019)    Acceptance Rate: 47.4%
Journals
  • F. Nussbaum and J. Giesen (2020). Pairwise sparse+ low-rank models for variables of mixed type. Journal of Multivariate Analysis, 178, 104601.

Software

  • model selection for conditional Gaussian (CG) distributions: cgmodsel

Education