Robert Gens

Robert Gens

I am a PhD student in UW CSE. My goal is to find architectural principles that will allow computers to understand the megapixels of our visual world as rapidly as we do. My advisor is Pedro Domingos. I received my BS in Electrical Engineering and Computer Science from MIT in 2009. In 2014 I will be supported by the Google PhD Fellowship in Deep Learning.


  • Learning the Structure of Sum-Product Networks. [pdf] [slides] [code and supplemental results]
    Robert Gens and Pedro Domingos.
    International Conference on Machine Learning 30 (ICML 2013).
  • Discriminative Learning of Sum-Product Networks. [pdf] [slides] [bib] [Talk Video]
    Robert Gens and Pedro Domingos.
    Advances in Neural Information Processing Systems 25 (NIPS 2012).
    Outstanding Student Paper Award

Workshop Papers

  • Learning Selective Sum-Product Networks. [pdf]
    Robert Peharz, Robert Gens, and Pedro Domingos.
    ICML 2014 Workshop on Learning Tractable Probabilistic Models, digital.
  • Competitive Learning for Deep Temporal Networks. [pdf]
    Robert Gens and Pedro Domingos.
    NIPS 2009 Workshop on Deep Learning for Speech Recognition and Related Applications, digital.


Commentary on "Learning Deep Architectures for AI" by Yoshua Bengio. As part of the CSE 590A (Winter 2010) Machine Learning reading group.
Commentary on "Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations" by Honglak Lee et al. . As part of the CSE 590M2 (Autumn 2009) Machine Learning reading group.