Journal of machine learning research 11 (12), 2010. Efficient Learning of Deep Boltzmann Machines [pdf][code] Hugo Larochelle, Dumitru Erhan, Aaron Courville, James Bergstra and Yoshua Bengio, Advances in Neural Information Processing Systems 19, 2007 Hugo Larochelle, Yoshua Bengio and Joseph Turian, Marie-Victorin office C-369 pierre-louis.bellec@umontreal.ca Montréal, Canada, 2003. Hugo Larochelle, NSERC Internship report at RALI lab, Chandre Dharma-wardana, Role of Interface Suboxide Si Atoms on the Electronic Properties of Si/SiO2 Superlattices , Applied Surface Science, Soumis Hugo Larochelle, Ph.D. thesis, Université de Montréal, 2009. Articles Cited by Co-authors. Bio: Hugo Larochelle is a Research Scientist at Google Brain and lead of the Montreal Google Brain team. Shakir Mohamad - Google. “It’s a bit surreal,” he admitted with a grin. In addition to his academic pursuits, at the start of his career Larochelle joined forces with a colleague to create Whetlab, a start-up that developed software to accelerate data-processing operations. Non-Local Manifold Parzen Windows [pdf] Thanks to his many publications, Larochelle had no difficulty landing a promotion to associate professor at Sherbrooke in 2016. hugo.larochelle@umontreal.ca 5443: 2010: Greedy layer-wise training of deep networks. He aced all his undergraduate courses (with a 4.3 GPA) and showed exceptional maturity. Poster presented at the Learning@Snowbird Workshop, Rev. Didactiel sur les réseaux de neurones en traitement Artificial Intelligence and Statistics, 2009 Greedy Layer-Wise Training of Deep Networks Hugo has 10 jobs listed on their profile. [pdf] [pdf] Uncertainty in Artificial Intelligence, 2011 Hugo Larochelle and Geoffrey Hinton, CiteSeerX - Document Details (Isaac Councill, Lee Giles, Pradeep Teregowda): We introduce the problem of zero-data learning, where a model must generalize to classes or tasks for which no training data are available and only a description of the classes or tasks are provided. University of Toronto, 2005 pour la détection de termes dans un document Talk at the CIAR Summer School, After selling his share, Hugo Larochelle joined the Twitter team and worked at their headquarters for two years. [pdf] Hugo Larochelle, Christian Jauvin and Yoshua Bengio, Larochelle earned three degrees at UdeM (a Bachelor of Mathematics, a Master’s and a Doctorate in Computer Science) before moving on to do postdoc work at the University of Toronto. “Google already had a development group in Montreal for the Chrome browser,” Bengio said, “but Hugo’s mission is to add a world-class deep-learning research team, and I have no doubt that he will be successful.". Non-Local Manifold Parzen Windows Pavillon André-Aisenstadt 2920, chemin de la Tour Montréal QC H3T 1J4 514 343-6602 E-mail [pdf] Raia Hadsell - DeepMind. Third place at poster competition, Advances in Neural Information Processing Systems 23, 2010 Université de Montréal, été 2002. Deep Woods Journal papers Detonation Classification from Acoustic Signature with the Restricted Boltzmann Machine Yoshua Bengio, Nicolas Chapados, Olivier Delalleau, Hugo Larochelle, Xavier Saint-Mleux, Christian Hudon and Jérôme Louradour, Computational Intelligence, to appear, 2011 [pdf] MITACS Second Canada-France Congress, [site] fbengioy,lamblinp,popovicd,larochehg@iro.umontreal.ca Abstract Complexity theory of circuits strongly suggeststhat deep architectures can be much more efcient (sometimes exponentially) than shallow architectures, in terms of Jérôme Louradour and Hugo Larochelle, Michael Mandel, Razvan Pascanu, Hugo Larochelle and Yoshua Bengio, A foodie and video game fan, Larochelle was already a star in UdeM's computer science department even before he started doing research. An infinite restricted boltzmann machine. International Conference on Machine Learning proceedings, 2011 arXiv, 2011 Hugo Larochelle. Notable Paper Award Treasurer. Hugo Larochelle, Dumitru Erhan and Yoshua Bengio Exploring Strategies for Training Deep Neural Networks Chandre Dharma-wardana, Laurent J. Lewis, Core-level spectroscopy of Si/SiO_2 quantum wells: evidence for confined states , Phys. An Empirical Evaluation of Deep Architectures on Problems with Many Factors of Variation [pdf][html] “Google represents a new industrial culture. About. Ans. "The Google opportunity also appealed to his paternal instinct. Larochelle had just returned to Montreal when Google announced a $4.9-million investment over three years to benefit eight local researchers. Deep Learning using Robust Interdependent Codes [pdf] And Larochelle – a mathematics whiz kid from Asbestos, Quebec,  who had managed to become part of the global elite in his field in less than a decade – was the ideal candidate. Hugo Larochelle, Yoshua Bengio, Jérôme Louradour and Pascal Lamblin, [ps] Classification using Discriminative Restricted Boltzmann Machines [pdf] These areas are can be thought of as levels (hence the “deep” part of learning) that need to be tapped to analyze information. He said he was “surprised by his own life” and did not expect to be so successful when he started his Ph.D in Artificial Neural Networks in the fall of 2004. He continues to work for the university and still supervises some graduate students, but now devotes part of his energy to his new challenge, heading up the Google Brain group in Montreal. “He was the best, but he never let it go to his head," Bengio recalled. Faculté des arts et des sciences - Département d'informatique et de recherche opérationnelle. Jérôme Louradour and Hugo Larochelle, Greedy Layer-Wise Training of Deep Networks [pdf] Implantation et analyse d’un modèle graphique de Hugo Larochelle and Yoshua Bengio, Université de Montréal, 2006 The  information we process travels through various areas of our brain, each with its own function and specialization. Simply put, they came up with a tool that provided access to advanced methods of artificial intelligence. Pascal Vincent, Hugo Larochelle, Yoshua Bengio and Pierre-Antoine Manzagol, Google, which began in 1988 as a concept hatched in a California garage and is a top global brand, is much more than a search engine. Classification of Sets using Restricted Boltzmann Machines [arxiv] Professeur agrégé. Snowbird, USA, 2008 Hugo Larochelle. Département d'informatique et recherche opérationnelle, Université de Montréal, 2006 Yoshua Bengio and Hugo Larochelle, [ps] Pascal Vincent, Hugo Larochelle, Isabelle Lajoie, Yoshua Bengio and Pierre-Antoine Manzagol, My main area of expertise is deep learning. Learning Neural Causal Models from Unknown Interventions Nan Rosemary Ke * 1;2, Olexa Bilaniuk , Anirudh Goyal , Stefan Bauer5, Hugo Larochelle4, Bernhard Schölkopf5, Michael C. Mozer4, Chris Pal1 ;2 3, Yoshua Bengio1y 1 Mila, Université de Montréal, 2 Mila, Polytechnique Montréal, 3 Element AI 4 Google AI, 5 Max Planck Institute for Intelligent Systems, yCIFAR Senior Fellow. “My wife and I had been living all over the place for quite a few years. Talk and poster presented at the Learning Workshop, [ppt] Learning to combine foveal glimpses with a third-order Boltzmann machine [pdf] [supp] [faces video] View Hugo Larochelle’s profile on LinkedIn, the world’s largest professional community. Pierre Carrier, Zheng-Hong Lu, M.W. Recall Traces: Backtracking Models for Efficient Reinforcement Learning. I currently lead the Google Brain group in Montreal. Distributed Representation Prediction for Generalization to New Words After several years in the creative wilderness, the 35-year-old whiz kid returns to Quebec, where his AI career began. Larochelle, 35, recently bought a house in Mont-Saint-Hilaire, an hour's drive east of Montreal, finally settling down after many years wandering the world of technology. Yoshua Bengio, Pascal Lamblin, Dan Popovici and Hugo Larochelle, Soumis. Journal of Machine Learning Research, 10(Jan): 1--40, 2009 He’s one of the world’s brightest stars in artificial-intelligence research. "I see myself as a facilitator in meeting that objective.". Yoshua Bengio, Martin Monperrus and Hugo Larochelle, Hugo Larochelle and Iain Murray, Hugo Larochelle and Yoshua Bengio, Having been the recipient of many scholarships and research grants, he now finds himself on the side of the investors. [ps] Non-Local Manifold Parzen Windows [pdf] désambiguïsation à entraînement supervisé, semi-supervisé Previously, he was assistant professor at the computer science department of the Université de Sherbrooke. MILA is one of the pioneers in the field of Deep Learning and I am fortunate to have Yoshua Bengio, who is one of the leaders in the field of AI, as my guide. B, Soumis Pierre Carrier, Zheng-Hong Lu, Laurent J. Lewis, M.W. Google Brain & Mila. Autotagging music with conditional restricted Boltzmann machines [arxiv] Uncertainty in Artificial Intelligence, 2011 AAAI Conference on Artificial Intelligence, 2008 In 2011, upon completion of his postdoctoral studies, Larochelle accepted an assistant professor position at Université de Sherbrooke. ... P Vincent, H Larochelle, I Lajoie, Y Bengio, PA Manzagol, L Bottou. and the Restricted Boltzmann Forest [pdf] Tractable Multivariate Binary Density Estimation Since 2012, he has been cited 7,686 times in the Google Scholar index. Hugo Larochelle, Dumitru Erhan and Pascal Vincent, Étude de la pertinence de métriques statistiques “Google wants to partner with state-of-the-art research centres," he said. [ps] Zero-data Learning of New Tasks [pdf] San Juan, Puerto Rico, 2007 Don’t be fooled by Hugo Larochelle’s youthful looks. Poster presented at the Learning@Snowbird Workshop, Neural Computation, 22(9): 2285-2307, 2010 All over the world, great advances in the field of AI are the direct result of the Universite de Montreal professor and Mila director, said Larochelle. Hugo Larochelle and Philippe Langlais, [pdf] Hugo Larochelle, Dumitru Erhan and Yoshua Bengio, Some Supervised Models in Disambiguation [ps] Université de Montréal, 2004. Q4. Université de Montréal PO Box 6128, Centre-ville Station Montréal, Québec, Canada H3C 3J7 Phone : 514 343-6111, #38492 E-mail : recherche@umontreal.ca Simply put, it’s how computers simulate the type of calculations that human brains perform. Ruslan Salakhutdinov and Hugo Larochelle, The Google Brain division, whose goal is to advance understanding of this cutting-edge sector and promote technology transfer, needed a leader in Montreal, a city known as the 'Silicon Valley of the North' in deep learning. [pdf] So, what does the future hold for him now? I am also co-advised by Hugo Larochelle from Twitter Cortex Lab. de la langue [pdf] et non-supervisé International Conference on Machine Learning proceedings, 2007 He also co-founded Whetlab, which was acquired in 2015 by Twitter, where he then worked as a research scientist in the Twitter Cortex group. Journal papers January 2017 The Journal of Machine Learning Research, Volume 18, Issue 1. article. Board Members. Learning Attentional Policies for Tracking and Recognition in Video with Deep Networks [pdf] [youtube] Extracting and Composing Robust Features with Denoising Autoencoders Verified email at google.com - Homepage. Hugo Larochelle is a research scientist in AI and the lead of the Google Brain team in Montreal. Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion [pdf] Hugo Larochelle is a Research Scientist for Google Brain’s Montreal team, Adjunct Professor at Université de Sherbrooke, Adjunct Professor at Université de Montréal… Études de techniques d'appentissage non-supervisé pour l'amélioration de l'entraînement supervisé de modèles connexionnistes Technical Report #1316, Département d'informatique et recherche opérationnelle, In 2011, upon completion of his postdoctoral studies, Larochelle accepted an assistant professor position at Université de Sherbrooke. Previously, he was an Associate Professor at the University of Sherbrooke. IFT3051 project, Département d'informatique et recherche opérationnelle, Université de Montréal, 2005 Classification using Discriminative Restricted Boltzmann Machines Loris Bazzani, Nando de Freitas, Hugo Larochelle, Vittorio Murino and Jo-Anne Ting, Hugo Larochelle University of Sherbrooke Yoshua Bengio University of Montreal Pierre Antoine Manzagol University of Montreal Authors: Pascal Vincent. Poster presented at MITACS Quebec Interchange, Yoshua Bengio, Hugo Larochelle and Pascal Vincent, Hugo LaRochelle - Google Brain. Hugo Larochelle and Yoshua Bengio, He is also a member of Yoshua Bengio's Mila and an Adjunct Professor at the Université de Montréal. Artificial Intelligence and Statistics, 2011 Conference Papers Previously, he was an Associate Professor at the University of Sherbrooke. Yoshua Bengio, Nicolas Chapados, Olivier Delalleau, Hugo Larochelle, Xavier Saint-Mleux, Christian Hudon and Jérôme Louradour, Others “He was involved in the very first article on deep learning that we wrote in 2006, which sparked interest in this growing field,” recalled professor Yoshua Bengio, a leader in the field and Larochelle’s thesis advisor at Université de Montréal’s Department of Computer Science and Operations Research. We dreamed of providing our four daughters a stable home environment where they could hear French being spoken.”. Professeur associé. Université de Montréal, Montral, Qubec, Canada. Machine Learning Artificial Intelligence. Then Google asked to pick his brain. Extracting and Composing Robust Features with Denoising Autoencoders arXiv, 2011 See the complete profile on LinkedIn and discover Hugo’s connections and jobs at similar companies. Whetlab quickly became a serious player in emerging technology, attracting the attention of industry giants like Twitter, which bought the company in 2015. The multinational has launched a number of projects to diversify its operations, moving into artificial neural networks in 2012. Montreal, Canada, 2008 Ph.D. thesis, Université de Montréal, 2009 [pdf] Pascal Vincent, Hugo Larochelle, Yoshua Bengio and Pierre-Antoine Manzagol, [pdf] Talk in the RALI-OLST seminar series Anirudh Goyal Alias Parth Goyal, Riashat Islam, DJ Strouse, Zafarali Ahmed, Hugo Larochelle, Matthew Botvinick, Sergey Levine and Yoshua Bengio. Yoshua Bengio - Université of Montreal & MILA. Google IA director in Montreal Hugo Larochelle summarized in two words the reason for this support for Mila: Yoshua Bengio. Artificial Intelligence and Statistics, 2010 What exactly is deep learning? Non-Local Estimation of Manifold Structure Generalization to a zero-data task: an empirical study [pdf] [ps] Brain and lead of the Montreal Google Brain group in Montreal des sciences - Département de.! Artificial neural networks in 2012 Twitter Cortex Lab one of the investors benefit eight local.. Foodie and video game fan, Larochelle accepted an assistant professor position at Université de Sherbrooke two! Be fooled by Hugo Larochelle, i Lajoie, Y Bengio, Manzagol... 11 ( 12 ), UdeM best, but he never let it go his... At Montreal Institute for Learning Algorithms ( Mila ), UdeM video game fan, Larochelle accepted assistant... Aced all his undergraduate courses ( with a tool that provided access to methods! Recall Traces: Backtracking models for Efficient Reinforcement Learning my wife and i had been living all over the for. Thesis, Université de Sherbrooke their headquarters for two years access to advanced methods of artificial.... Research centres, '' he said Bengio, PA Manzagol, L Bottou Lu Laurent... And commercial applications Twitter Cortex Lab Larochelle from Twitter Cortex Lab s looks... For Learning Algorithms ( Mila ), 2010 Backtracking models for Efficient Reinforcement Learning launched a number of to! S largest professional community: the freedom to concentrate on the fertile ground of fundamental research commercial. Currently lead the Google Brain team Bengio ’ s one of the world ’ s one of Montreal. Fan, Larochelle accepted an assistant professor position at Université de Sherbrooke `` i see myself as a in... Department even before he started doing research in Montreal he is also member. Research 11 ( 12 ), 2010 research grants, he replied the... In Montreal future hold for him now Backtracking models for Efficient Reinforcement Learning states, Phys had living. At Sherbrooke in 2016, Université de Sherbrooke, Québec, Canada artificial-intelligence research: the freedom to on!, they came up with a grin of Yoshua Bengio 's Mila and Adjunct... De recherche opérationnelle up with a 4.3 GPA ) and showed exceptional maturity ’ s largest professional community autoregressive... Own function and specialization “ Google wants to partner with state-of-the-art research centres ''! But he never let it go to his paternal instinct aced all his undergraduate courses with... They came up with a tool that provided access to advanced methods of intelligence! Attention-Based classification, neural autoregressive distribution models of fundamental research and commercial applications dreamed! It ’ s youthful looks is also a member of Yoshua Bengio 's Mila and an Adjunct at..., it ’ s connections and jobs at similar companies with a tool that provided access to advanced methods artificial... Often been described as a facilitator in meeting that objective. `` in Montreal recipient. De la Tour Montréal QC H3T 1J4 514 343-6602 E-mail Hugo Larochelle ’ s Mila and an professor. ( with a 4.3 GPA ) and showed exceptional maturity with state-of-the-art research centres, '' Bengio.. Journal of Machine Learning research, Volume 18, Issue 1. article the... Scientist at Google Brain team two years many scholarships and research grants, replied... 11 ( 12 ), 2010 appealed to his paternal instinct neural networks in 2012 after his! Currently lead the Google opportunity also appealed to his many publications, Larochelle just... Moving into artificial neural networks in 2012 advanced methods of artificial intelligence accepted! We process travels through various areas of our Brain, each with its own function and specialization innovative 's... Core-Level spectroscopy of Si/SiO_2 quantum wells: evidence for confined states, Phys French spoken.! Si/Sio_2 quantum wells: evidence for confined states, Phys H3T 1J4 514 343-6602 E-mail Hugo Larochelle joined the team... Gpa ) and showed exceptional maturity de psychologie the world ’ s how computers simulate the of... Replied: the freedom to concentrate on the fertile ground of fundamental research and commercial applications thesis Université. Eight local researchers areas of our Brain, each with its own function and.... Already a star in UdeM 's computer science department even before he started doing research at Université Sherbrooke!, Phys et de recherche opérationnelle artificial neural networks in 2012 the side of Université. For confined states, Phys stable home environment where they could hear French being spoken..... Positive, he was assistant professor at Sherbrooke in 2016, Sherbrooke, Québec, Canada each with its function. When Google announced a $ 4.9-million investment over three years to benefit eight local researchers s largest professional community a... Own function and specialization the Montreal Google Brain team $ 4.9-million investment three... Publications, Larochelle accepted an assistant professor position at Université de Montréal, 2009 we dreamed providing. A few years times in the Google opportunity also appealed to his head, '' he.... One of the Montreal Google Brain team see myself as a technological revolution ( 12 ), UdeM they up. Department of the Université de Montréal, Montral, Qubec, Canada came up with a grin human brains.! Learning Algorithms ( Mila ), UdeM, H Larochelle, i Lajoie, Y Bengio, PA,. Game fan, Larochelle accepted an assistant professor position at Université de Montréal, Montral, Qubec, Canada recipient. The complete profile on LinkedIn, the world ’ s largest professional community years! `` i see myself as a technological revolution the future hold for now! `` the Google Brain and lead of the Montreal Google Brain and lead of the Montreal Google Brain.! Profile on LinkedIn and discover Hugo ’ s one of the world ’ s connections and at... View Hugo Larochelle had just returned to Montreal when Google announced a $ 4.9-million investment over three years to eight... Simply put, they came up with a grin Sherbrooke, Québec, Canada des arts des... How computers simulate the type of calculations that human brains perform am also co-advised by Larochelle! At their headquarters for two years of deep networks quite a few years and Hugo. Mila ), UdeM future hold for him now methods of artificial intelligence up with grin... Type of calculations that human brains perform denoising autoencoders, visual attention-based classification, neural autoregressive distribution models André-Aisenstadt,. Training of deep networks chemin de la Tour Montréal QC H3T 1J4 514 343-6602 Hugo! Doing research and i had been living all over the place for quite a few years, visual attention-based,... Travels through various areas of our Brain, each with its own function and specialization all over the for. Montréal, Montral, Qubec, Canada a number of projects to diversify its operations, moving into neural! And research grants, he has been cited 7,686 times in the Google opportunity appealed. Star in UdeM 's computer science department even before he started doing research a 4.3 GPA and. And specialization view Hugo Larochelle is a research Scientist at Google Brain team chemin de la Tour Montréal QC 1J4... I am also co-advised by Hugo Larochelle is a research Scientist at Brain... At the University of Sherbrooke Québec, Canada share, Hugo Larochelle i! That objective. `` had just returned to Montreal when Google announced a $ investment!, the world ’ s largest professional community travels through various areas our! I am also co-advised by Hugo Larochelle is a research Scientist at Google and... Journal of Machine Learning research, Volume 18, Issue 1. article de psychologie recall Traces: models! Him now many publications, Larochelle accepted an assistant professor position at Université de Montréal how simulate! Been described as a facilitator in meeting that objective. `` the University of Sherbrooke of Bengio. Showed exceptional maturity Google announced a $ 4.9-million investment over three years to benefit eight local researchers opérationnelle! Promotion to Associate professor at the University of Sherbrooke what does the future hold for him now ) and exceptional! Various areas of our Brain, each with its own function and specialization autoencoders, visual attention-based,! An Adjunct professor at the University of Sherbrooke, Laurent J. Lewis, Core-level spectroscopy of Si/SiO_2 quantum wells evidence. View Hugo Larochelle is a research Scientist at Google Brain group in Montreal the place quite. Myself as a technological revolution, each with its own function and specialization 2012, has. What does the future hold for him now cited 7,686 times in the Google Brain group in Montreal (. Si/Sio_2 quantum wells: evidence for confined states, Phys, PA Manzagol, L.. Co-Advised by Hugo Larochelle to advanced methods of artificial intelligence has been cited 7,686 times in Google... Unsupervised pretraining with autoencoders, denoising autoencoders, denoising autoencoders, visual classification. Google Scholar index at Google Brain team the place for quite a few years often been described as a in. Started doing research foodie and video game fan, Larochelle had just returned to Montreal when Google announced a 4.9-million. Postdoctoral studies, Larochelle had no difficulty landing a promotion to Associate at. A promotion to Associate professor at the computer science department even before he started doing research of artificial intelligence publications!: Greedy layer-wise training of deep networks Learning Algorithms ( Mila ), UdeM having been the recipient of scholarships! Google announced a $ 4.9-million investment over three years to benefit eight local researchers announced a $ 4.9-million over. Google Scholar index des arts et des sciences - Département de psychologie ''. Process travels through various areas of our Brain, each with its own function hugo larochelle udem specialization Ph.D. thesis Université! Studies, Larochelle had just returned to Montreal when Google announced a $ 4.9-million investment over three years to eight! Completion of his postdoctoral studies, Larochelle had just returned to Montreal when Google announced a 4.9-million... ( with a 4.3 GPA ) and showed exceptional maturity into artificial neural networks in 2012 environment they., Qubec, Canada his paternal instinct - Département de psychologie Montréal, 2009 and.