About me

I'm a PhD candidate at the University of Montreal supervised by Aaron Courville, and a member of the MILA lab headed by Yoshua Bengio. My current research is focused on language grounding, visual reasoning and goal-oriented dialogue systems. In the past, I have also worked on optimization methods for deep neural networks.


Publications

Learning visual reasoning without strong priors
Ethan Perez, Harm de Vries, Florian Strub, Vincent Dumoulin and Aaron Courville

Modulating early visual processing by language
Harm de Vries*, Florian Strub*, Jérémie Mary, Hugo Larochelle, Olivier Pietquin and Aaron Courville
NIPS 2017 (Spotlight)
[ArXiv]

End-to-end optimization of visually grounded dialogue systems
Florian Strub, Harm de Vries, Jérémie Mary, Bilal Piot, Aaron Courville and Olivier Pietquin
IJCAI 2017
[arXiv], [code]

GuessWhat?! Visual object discocery through multi-modal dialogue
Harm de Vries, Florian Strub, Sarath Chandar, Olivier Pietquin, Hugo Larochelle and Aaron Courville
CVPR 2017 (Spotlight)
[arXiv], [code]

Deep Learning Vector Quantization
H. de Vries, R. Memisevic and A. Courville
European Symposium on Artificial Neural Networks 2016 (Oral)

Equilibrated adaptive learning rates for non-convex optimization
Y. Dauphin*, H. de Vries* and Y. Bengio
NIPS 2015 (Spotlight)
[arXiv], [NIPS link]

Can deep learning help you find the perfect match?
H. de Vries and J. Yosinski
Deep learning work shop, ICML 2015
[arXiv], [Blog post]

Parametric nonlinear regression models for dike monitoring
H. de Vries, G. Azzopardi, A. Koelewijn, A. Knobbe
Advances in Intelligent Data Analysis 2014 (Oral)
[Link], [Pdf]

Stationarity and uniqueness of Generalized Matrix Learning Vector Quantization
H. de Vries
Mittweida Workshop on Computational Intelligence 2013
[Link]