Florian Wenzel

Humboldt University of Berlin
Department of Computer Science

Rudower Chaussee 25
Building Section 4, Room 217
12489 Berlin

wenzelfl@hu-berlin.de

Short Bio

Since October 2015, I am a PhD student in the machine learning research group at Humboldt University of Berlin headed by Marius Kloft. I did a scientific internship at Disney Research in Pittsburgh (USA) from February to May 2017. In September 2015, I received a Master's degree (M.Sc.) in mathematics from the Humboldt University of Berlin. I wrote my Master's thesis in the field of probabilistic machine learning.

Research Interests

I am interested in probabilistic machine learning and its applications. In particular, I like:

  • Probabilistic Modeling
  • Scalable Bayesian Inference
  • Probabilistic Programming Languages
  • Gaussian Processes

News

  • Our paper Sparse Probit Linear Mixed Model got accepted in Machine Learning. You can check it out here.
  • Our paper Bayesian Nonlinear Support Vector Machines for Big Data got accepted as an oral at ECML and was nominated for best studen paper award. Here is the paper.
Last update 25/19/17.

Teaching


Past Courses:

Publications

2017
  • S. Mandt*, F. Wenzel*, S. Nakajima, J. P. Cunningham, C. Lippert, and M. Kloft: Sparse Probit Linear Mixed Model. Machine Learning Journal, Jul 2017, ISSN 1573-0565 . [PDF] [CODE]
    * = equal contribution
  • F. Wenzel, M. Deutsch, T. Galy-Fajou and M. Kloft: Bayesian Nonlinear Support Vector Machines for Big Data. ECML PKDD, 2017. (Best Student Paper Award Nomination / Oral Presentation) [PDF] [CODE]
2016
  • F. Wenzel, M. Deutsch, T. Galy-Fajou and M. Kloft: Scalable Approximate Inference for the Bayesian Nonlinear Support Vector Machine. Proceedings of the Workshop on Advances in Approximate Bayesian Inference at NIPS, 2016. [PDF]
  • P. Jähnichen, F. Wenzel and M. Kloft: Scalable Inference in Dynamic Mixture Models. Proceedings of the Workshop on Advances in Approximate Bayesian Inference at NIPS, 2016. [PDF]
  • P. Jähnichen, F. Wenzel, and M. Kloft: Scalable Inference in Dynamic Admixture Models. Proceedings of LWDA, 2016. (Oral Presentation)[PDF]
  • S. Mandt*, F. Wenzel*, S. Nakajima, C. Lippert, and M. Kloft: Separating Sparse Signals from Correlated Noise in Binary Classification. Proceedings of the UAI Workshop Causation: Foundation to Application, 2016. (Oral Presentation) [PDF]
    * = equal contribution
2015 and before
  • S. Mandt, F. Wenzel, S. Nakajima, J. P. Cunningham, C. Lippert, and M. Kloft: Finding Sparse Features in Strongly Confounded Medical Binary Data. NIPS 2015 Workshop on Machine Learning in Healthcare, 2015. (Oral Presentation) [PDF]
  • F. Wenzel: Probit Regression with Correlated Label Noise. Master's thesis, Humboldt University of Berlin, 2015. [PDF]
  • S. Mandt, F. Wenzel, J. Cunningham, and M. Kloft: Probit Regression with Correlated Label Noise: An EM-EP approach. Proceedings of the NIPS workshop on Advances in Variational Inference, 2014. [PDF]

Activities