martint.blog

photo

My name is Martin Trapp, I’m a PhD candidate in machine learning. I have a particular interested in deep tractable probabilistic models, probabilistic programming and on Bayesian nonparametrics for probabilistic modelling.

My thesis is supervised by Prof. Franz Pernkopf from the Graz University of Technology in cooperation with Dr. Robert Peharz from the University of Cambridge. I’m also affiliated to the OFAI led by Prof. Robert Trappl.

In my spare time, I’m working on the universal probabilistic programming language Turing.jl. Our goal is to provide efficient, robust and intuitive probabilistic programming for everyone.

To contact me, find me on Twitter or Github or drop me an e-mail.

News

  • [September 2019] Our paper on Bayesian Learning of Sum-Product Networks with Robert Peharz, Hong Ge, Franz Pernkopf and Zoubin Ghahramani has been accepted at NeurIPS (acceptance rate 21.2%).
  • [June 2019] I presented our work Turing.jl for Probabilistic Programming with Discrete Random Measures with Emile Mathiu, Maria Lomeli and Hong Ge at the International Conference on Bayesian Nonparametrics.
  • [May 2019] Our paper on Random Sum-Product Networks: A Simple and Effective Approach to Probabilistic Deep Learning with Robert Peharz, Antonio Vergari, Karl Stelzner, Alejandro Molina, Xiaoting Shao, Kristian Kersting, and Zoubin Ghahramani has been accepted at UAI (acceptance rate 26.2%).
  • [May 2019] Our paper on Optimisation of Overparametrized Sum-Product Networks with Robert Peharz and Franz Pernkopf has been accepted at the 3rd Workshop on Tractable Probabilistic Models at ICML
  • [June 2018] Our work on Learning Deep Mixtures of Gaussian Process Experts using Sum-Product Networks with Robert Peharz, Franz Pernkopf and Carl E. Rasmussen has been accepted at the 2nd Workshop on Tractable Probabilistic Models at ICML
  • [December 2017] I’ll be visiting the machine learning group at the University of Cambridge as a visiting scholar for two month.
  • [June 2017] I presented our work on Infinite Sum-Product Trees at the International Conference on Bayesian Nonparametrics.
  • [June 2017] Our paper on Safe Semi-Supervised Learning of Sum-Product Networks with Tamas Madl, Robert Peharz, Franz Pernkopf, and Robert Trappl has been accepted at UAI (acceptance rate 31.0%).