http://www.cs.toronto.edu/~larocheh/

Hugo Larochelle Postdoctoral Fellow Member of the machine learning group Department of Computer Science University of Toronto larocheh [at] cs [dot] toronto [dot] edu

I'm interested in machine learning algorithms, which are algorithms capable of extracting concepts and abstractions from data.

My current topic of interest is how to use unsupervised learning methods to improve the performance of models on supervised problems for which labeled data are available. In particular, I'm interested in applying these models on problems where the input data is high dimensional, like for images, texts or sounds.

Research projects

Here is an informal list of projects on which I'm working:

• Development of variants on Restricted Boltzmann Machines that improve their performance on supervised problems;
  
  
• Development and study of efficient training methods for deep neural networks with several hidden layers;
  
  

I graduated from University of Montreal in 2004, after completing a bidisciplinary program in Mathematics and Computer Science. During my studies, I worked as an intern in the RALI and LISA labs successively, as part of NSERC Undergraduate Student Research Awards. I also received the "Bourse du doyen de la Faculté des Arts et Sciences de l'Université de Montréal" two times.

I also recently obtained a Ph.D. under the supervision of professor Yoshua Bengio, also at University of Montreal. I received the Canada Graduate Master's Scholarship and the Canada Graduate Doctoral Scholarship from NSERC.