I am a fifth year PhD student in computer science at Mila, Université de Montréal, under the supervision of Pascal Vincent, specializing in machine learning, most specifically deep learning. I am strongly interested by both research and software development.
During my Master and PhD I did research on a wide array of subjects ranging from NLP, Computer Vision, optimization to theoretical deep learning. I am now concentrating on studying the methodology for deep learning research, designing methods to improve reproducibility.
Beside research, I have been an important contributor to tools developed at LISA/Mila such as Theano and Pylearn2 and a mentor for students contributing to these projects. I am now the lead developer of the project Oríon, a platform for hyper-parameter optimization, supported by the Mila IDT team and for which we are currently signing collaboration agreements with the industry. I am concurrently developing an automation tool called Mahler, which enables large scale experimentations for improved reproducibility using statistical tests.