Lisa_seminaires Janvier 2017

lisa_seminaires@iro.umontreal.ca

3 participants
9 discussions

[TeaTalk] Prof. Hervé Lombaert, Feb 3, 13:30, AA6214
by Dzmitry Bahdanau 03 Fév '17

03 Fév '17

Hi all, You are invited to attend the talk by Prof. Hervé Lombaert this Friday, Feb 3, at 13:30, room AA6214. *Title:* Spectral Matching & Learning of Surface Data - Example on Brain Surfaces *Abstract:* How to analyze complex shapes, such as of the highly folded surface of the brain? In this talk, I will show how spectral representations of shapes can benefit learning problems where data lives on surfaces. Key operations, such as segmentation and registration, typically need a common mapping of surfaces, often obtained via slow and complex mesh deformations in a Euclidean space. Here, we exploit spectral coordinates derived from the Laplacian eigenfunctions of shapes and also address the inherent instability of spectral shape decompositions. Spectral coordinates have the advantage over Euclidean coordinates, to be geometry aware and to parameterize surfaces explicitly. This change of paradigm, from Euclidean to spectral representations, enables a classifier to be applied *directly* on surface data, via spectral coordinates. The talk will focus, first, on spectral representations of shapes, with an example on brain surface matching, and second, on the learning of surface data, with an example on automatic brain surface parcellation. *Speaker's bio: *Hervé Lombaert is a Starting Research Scientist at Inria Sophia-Antipolis, France, and Associate Professor at ETS, Montreal - with research interests in Statistics on Shapes, Data & Medical Images. He had the chance to work in multiple centers, including Microsoft Research (Cambridge, UK), Siemens Corporate Research (Princeton, NJ), Inria Sophia-Antipolis (France), McGill University (Canada), and Polytechnique Montreal (Canada). He is also a recipient of the François Erbsmann Prize, a top prize in Medical Image Analysis, earned the select NSERC Postdoctoral Fellowship and the FQRNT Étudiant-Chercheur Étoile - more at [ http://cim.mcgill.ca/~lombaert] Best, Dima

1 1

Fwd: Colloque DIRO, *lundi* 30 janvier, François Fouquet (Luxembourg)
by Yoshua Bengio 26 Jan '17

26 Jan '17

Another job talk with some machine learning in it, more, hopefully. ---------- Forwarded message ---------- From: Pierre McKenzie <mckenzie(a)iro.umontreal.ca> Date: 2017-01-26 18:54 GMT-05:00 Subject: Colloque DIRO, *lundi* 30 janvier, François Fouquet (Luxembourg) To: seminaires(a)iro.umontreal.ca *Enabling Model-Driven Analytics for Cyber Physical systems * par * François Fouquet * University of Luxembourg *Lundi 30 janvier, 15:30-16:30*, *Salle 3195*, Pavillon André-Aisenstadt Université de Montréal, 2920 Chemin de la Tour Café avant 15:00-15:30 *Cette présentation sera donnée en français.* *Résumé:* Advances in software, embedded computing, sensors, and networking technologies will lead to a new generation of smart cyber-physical systems. Such systems have to continuously monitor and collect data from physical processes in order to make decisions fast, i.e., in live. However, the complex nature combined with the massive amount of data generated by such systems impose fundamental challenges. Smart grids are excellent illustrations of such systems. They combine novel country-wide communication medias together with many energy sensors to enable the next generation of sustainable infrastructure, including renewable energies. For instance, on-field grid operators can now leverage these newly collected data to empower live simulation tools that allow them to foresee the impacts of maintenance actions before their actual application on the real grid. However, developing such software systems, at the crossroad of big data, modelling and simulation is very challenging. In collaboration with the main electricity grid operator in Luxembourg, we have developed a novel analytic approach to process such complex data in live. During this talk, I will present what we call model-driven live analytics. I will focus on the key enablers to make this approach scalable to the size of country-wide CPSs. The main contribution of this work is a multi-dimensional graph data model that brings raw data, domain knowledge, and machine learning together in a single model, which can drive live analytic processes. Firstly, data handled by cyber-physical systems is usually dynamic and changes frequently and at different paces. I will present a temporal graph data model and storage system, which consider time as a first-class property and allow to analyse frequently changing data. Additionally, I will present how a continuous sequence of sensor values can be efficiently encoded using live mathematical model inference. Secondly, making sustainable decisions requires to anticipate which impacts certain actions could have. In some cases, hundreds or thousands of such hypothetical actions must be explored ahead before any solid decision can be taken. I will present our approach to deal with such need - a multi-dimensional model that efficiently represent, store and analyse many different alternatives of the same system in live. Thirdly, to make smart decisions, cyber-physical systems must continuously refine behavioural models that are known at design time, with what can only be learned from live data. During this talk, I will present how we have combined machine learning and the multi-dimensional graph data model to empower live analytics for cyber-physical systems. Finally, I will conclude this talk with details about the open source project, which is developed around these research ideas, and share the lessons learned with respect to high-performance Java and big data.

1 0

[TeaTalk] Talks by Çağlar Gülçehre and Vincent Dumoulin
by Dzmitry Bahdanau 26 Jan '17

26 Jan '17

Hi all, Next Friday (27.01) at 1:30pm, room 6214, we will have not just one, but two tea-talk presentations by Çağlar and Vincent, both MILA students. Please come in great numbers! Speaker 1: Çağlar Gülçehre Title: TARDIS: an RNN with Wormhole Connections Abstract: In this talk, I will discuss a new memory augmented model which we have been working on with Sarath and Yoshua since last summer. Basically we revisit the idea of memory augmented models and investigate some of their aspects theoretically and propose a new model. The presentation will be about an ongoing work(although the arXiv paper will be out soon). Basically, we revisit the idea of memory augmented models and investigate some of their aspects theoretically and propose a new model. In this paper, we propose a novel memory augmented neural network model called TARDIS (Temporal Automatic Relation Discovery in Sequences). Our model (TARDIS) has an RNN controller which modifies an external memory component where it can store a selective set of embeddings of its own previous hidden states and revisit them as and when needed. In terms of the memory structure, our proposed model is similar to Dynamic Neural Turing Machines ~(D-NTM), but both read and write operations are simpler and more efficient. TARDIS uses discrete addressing for read/write operations which helps to substantially reduce the vanishing gradient problem with very long sequences. We evaluate our models on different long-term dependency tasks and report competitive results in all of them. Speaker 2: Vincent Dumoulin Title: A Learned Representation for Artistic Style Abstract: In this work (done while interning at Google Brain and submitted to ICLR2017) we investigate the construction of a single, scalable deep network that can parsimoniously capture the artistic style of a diversity of paintings. We demonstrate that such a network generalizes across a diversity of artistic styles by reducing a painting to a point in an embedding space. Importantly, this model permits a user to explore new painting styles by arbitrarily combining the styles learned from individual paintings. Best, Dima

1 1

Fwd: Rappel: jeudi 26 janvier (attention à la salle), Pooyan Fazli (Cleveland State)
by Yoshua Bengio 26 Jan '17

26 Jan '17

Another robotics job talk, 3:30pm, rm 6214. ---------- Forwarded message ---------- From: Pierre McKenzie <mckenzie(a)iro.umontreal.ca> Date: 2017-01-25 22:55 GMT-05:00 Subject: Rappel: jeudi 26 janvier (attention à la salle), Pooyan Fazli (Cleveland State) To: seminaires(a)iro.umontreal.ca Bienvenue à tous! -------- Forwarded Message -------- Subject: Colloque DIRO, jeudi 26 janvier, Pooyan Fazli (Cleveland State U.) Date: Fri, 20 Jan 2017 11:55:05 -0500 From: Pierre McKenzie <mckenzie(a)iro.umontreal.ca> <mckenzie(a)iro.umontreal.ca> To: seminaires(a)iro.umontreal.ca *Distributed Robotics: Algorithms and Applications * par * Pooyan Fazli * Cleveland State University *Jeudi 26 janvier, 15:30-16:30*, *Salle 6214*, Pavillon André-Aisenstadt Université de Montréal, 2920 Chemin de la Tour Café avant 15:00-15:30 *Cette présentation sera donnée en anglais.* *Résumé:* Future robots will have to work in dynamic, uncertain, adversarial, and multi-goal environments. Thus, the integration of the advanced physical and cognitive systems required by the next generation of robots is a challenging task. It is not feasible to design a ‘universal’ robot capable of working across a wide range of applications. The robots of today are examples of the challenges of cost, long training period, and limited functionality, which will apply to robots of the near future as well. Given these challenges, multi-robot systems may be more suitable than single-robot systems in many real world applications. In this talk, I shall discuss two challenging problems in distributed robotics: 1) Service robots frequently face similar tasks. However, they are still not able to share their knowledge efficiently on how to accomplish those tasks. We introduce a new framework, which allows remote and heterogeneous robots to share plans and instructions on the tasks assigned to them. We demonstrate the feasibility of the framework with experiments between two geographically distributed robots and analyze the performance of the framework quantitatively. 2) In distributed coverage, a team of robots is deployed to move around a target area to perform sensing, monitoring, data collection, search, or distributed servicing tasks. We focus on the problem of multi-robot coverage and patrolling of the boundaries of a target area and the structures inside it. Events may occur at any position on the boundaries, and the goal of the robot team is to maximize the total detection reward of the events. To this end, we introduce an online, distributed algorithm, in which each robot autonomously learns the event distribution on the boundaries and then plans accordingly to visit the most promising parts of the target area. We also investigate how robots’ visual range and communication among the robots affect the performance of the robot team in the coverage problem. *Biographie:* Pooyan Fazli is an assistant professor and the founding director of the People and Robots Laboratory (PeRL) in the Electrical Engineering and Computer Science Department at Cleveland State University. He received his Ph.D. in Computer Science from the University of British Columbia and his M.Sc. and B.Sc. in Computer Science from Amirkabir University of Technology. Prior to joining CSU, he was a postdoctoral fellow in the CORAL Research Group at Carnegie Mellon University and in the Laboratory for Computational Intelligence at the University of British Columbia. His research focuses on artificial intelligence, autonomous robots, multi-robot systems, human-robot teams/interaction, machine learning, and robot vision and perception.

1 0

Fwd: Postes ouverts à l'UQAM
by Yoshua Bengio 20 Jan '17

20 Jan '17

data science faculty positions at UQAM ---------- Message transféré ---------- De : Abdoulaye Banire Diallo <diallo.abdoulaye(a)uqam.ca> Date : 20 janvier 2017 à 11:01 Objet : Postes ouverts à l'UQAM À : Yoshua.Bengio(a)umontreal.ca Bonjour Yoshua, Nous avons deux postes de professeurs ouverts en sciences de données et systèmes distribués qui pourraient interesser des étudiants ou postdocs de chez vous. https://www.rhu.uqam.ca/AffichageProfs/21007007.pdf https://www.rhu.uqam.ca/AffichageProfs/21007008.pdf Bonne journée Aux plaisirs. *--* *Abdoulaye Baniré Diallo Ph.D.* *Directeur* de Maitrise et Doctorat en informatique *Directeur* du laboratoire de bioinformatique *Professeur permanent* *Membre régulier *de pharmaQAM *Université du Québec à Montréal* *Montréal, Qc, Canada* http://labo.bioinfo.uqam.ca

1 0

Fwd: Colloque DIRO, jeudi 26 janvier, Pooyan Fazli (Cleveland State U.)
by Yoshua Bengio 20 Jan '17

20 Jan '17

More robotics + machine learning next Thursday, another job talk. ---------- Forwarded message ---------- From: Pierre McKenzie <mckenzie(a)iro.umontreal.ca> Date: 2017-01-20 11:55 GMT-05:00 Subject: Colloque DIRO, jeudi 26 janvier, Pooyan Fazli (Cleveland State U.) To: seminaires(a)iro.umontreal.ca *Distributed Robotics: Algorithms and Applications * par * Pooyan Fazli * Cleveland State University *Jeudi 26 janvier, 15:30-16:30*, *Salle 6214*, Pavillon André-Aisenstadt Université de Montréal, 2920 Chemin de la Tour Café avant 15:00-15:30 *Cette présentation sera donnée en anglais.* *Résumé:* Future robots will have to work in dynamic, uncertain, adversarial, and multi-goal environments. Thus, the integration of the advanced physical and cognitive systems required by the next generation of robots is a challenging task. It is not feasible to design a ‘universal’ robot capable of working across a wide range of applications. The robots of today are examples of the challenges of cost, long training period, and limited functionality, which will apply to robots of the near future as well. Given these challenges, multi-robot systems may be more suitable than single-robot systems in many real world applications. In this talk, I shall discuss two challenging problems in distributed robotics: 1) Service robots frequently face similar tasks. However, they are still not able to share their knowledge efficiently on how to accomplish those tasks. We introduce a new framework, which allows remote and heterogeneous robots to share plans and instructions on the tasks assigned to them. We demonstrate the feasibility of the framework with experiments between two geographically distributed robots and analyze the performance of the framework quantitatively. 2) In distributed coverage, a team of robots is deployed to move around a target area to perform sensing, monitoring, data collection, search, or distributed servicing tasks. We focus on the problem of multi-robot coverage and patrolling of the boundaries of a target area and the structures inside it. Events may occur at any position on the boundaries, and the goal of the robot team is to maximize the total detection reward of the events. To this end, we introduce an online, distributed algorithm, in which each robot autonomously learns the event distribution on the boundaries and then plans accordingly to visit the most promising parts of the target area. We also investigate how robots’ visual range and communication among the robots affect the performance of the robot team in the coverage problem. *Biographie:* Pooyan Fazli is an assistant professor and the founding director of the People and Robots Laboratory (PeRL) in the Electrical Engineering and Computer Science Department at Cleveland State University. He received his Ph.D. in Computer Science from the University of British Columbia and his M.Sc. and B.Sc. in Computer Science from Amirkabir University of Technology. Prior to joining CSU, he was a postdoctoral fellow in the CORAL Research Group at Carnegie Mellon University and in the Laboratory for Computational Intelligence at the University of British Columbia. His research focuses on artificial intelligence, autonomous robots, multi-robot systems, human-robot teams/interaction, machine learning, and robot vision and perception.

1 0

[TeaTalk] Talk by Professor Brendan Frey on 'Using Machine Learning to Detect and Treat Genetic Disease'
by Junyoung Chung 20 Jan '17

20 Jan '17

Hi everyone, we will have our first tea talk on this Friday. Please NOTE that this tea talk is scheduled at **13:30-14:30** in **AA6214**. We are in a transition process of moving the tea talk time slot from 11:00-1200 to 13:30-14:30. Professor Brendan Frey from U of T will tell us about his work on applying machine learning to detect and treat genetic disease. If anyone is interested in meeting with Brendan, please tell Dima or me, we will host a short meeting after his talk. Who : Professor Brendan Frey When : 20/1/2017 13:30-14:30 Where: AA6214 Title: Using Machine Learning to Detect and Treat Genetic Disease Abstract: Diseases and corresponding treatments often have to do with genetic modifications, such as changing a T to a G at a particular location within DNA. In 2015, my University of Toronto group spun out a startup, Deep Genomics, which is using machine learning to model how these changes impact cellular processes involved in disease, such as cancer and neurological disorders. For example, to train our splicing model, we are using 9661 different datasets, each with a number of training cases that varies from 8 million to 220 million. In this talk, I'll describe some of the machine learning challenges that, if solved, will completely change the field of medicine, save hundreds of thousands of lives, and improve the quality of life for hundreds of millions of people. Biography: Brendan Frey is the CEO and Co-Founder of Deep Genomics and a Professor in Engineering and Computer Science at the University of Toronto. He has made fundamental contributions in the areas of machine learning and genome biology, and is widely known for his work on using machine learning to understand how genetic variation leads to disease. Dr. Frey has received numerous distinctions and is a Fellow of the Royal Society of Canada, a Senior Fellow of the Canadian Institute for Advanced Research, and a John C. Polanyi Fellow. He recently co-founded a University of Toronto spinoff, Deep Genomics, which is using machine learning to bring about a new approach to genomic medicine. Dr. Frey has consulted for several industrial research and development laboratories in Canada, the United States and England, and has served on the Technical Advisory Board of Microsoft Research. His former students and postdoctoral fellows include professors, industrial researchers and developers at universities and industrial laboratories from across Canada, the United States and Europe. Best, --Junyoung

1 1

Fwd: Rappel: jeudi 19 janvier, Lantao Liu (USC)
by Yoshua Bengio 18 Jan '17

18 Jan '17

Interesting seminar tomorrow by a potential new faculty hire in the area of robotics. ---------- Forwarded message ---------- From: Pierre McKenzie <mckenzie(a)iro.umontreal.ca> Date: 2017-01-18 21:51 GMT-05:00 Subject: Rappel: jeudi 19 janvier, Lantao Liu (USC) To: seminaires(a)iro.umontreal.ca Bienvenue à tous! -------- Forwarded Message -------- Subject: colloque DIRO, jeudi 19 janvier, Lantao Liu (USC) Date: Mon, 16 Jan 2017 14:20:32 -0500 From: Pierre McKenzie <mckenzie(a)iro.umontreal.ca> <mckenzie(a)iro.umontreal.ca> To: seminaires(a)iro.umontreal.ca * From Single-Robot Planning to Multi-Robot Collective Decision-Making* par *Lantao Liu * University of Southern California *Jeudi 19 janvier, 15:30-16:30*, *Salle 3195*, Pavillon André-Aidenstadt Université de Montréal, 2920 Chemin de la Tour Café avant 15:00-15:30 *Cette présentation sera donnée en anglais.* *Résumé:* Effectively understanding the surrounding environment and efficiently interacting with it are important functions for autonomous systems, where in a broad sense the environment consists of not only static objects but also dynamic things such as wind, fluid, and other robot teammates. The non-static environment also indicates that the underlying environmental model is time-variant, with possibly high uncertainty in both space and time dimensions. In this talk, I will first introduce a decision-making framework called time-varying Markov Decision Process, which can be used to cope with robot's time-varying action uncertainty caused by dynamic disturbances (e.g., fluid flows, air turbulences). Then I am going to describe a data-driven planning and learning method for long-term environment sensing and monitoring, where I use techniques of Gaussian Processes, information theory and Bayesian inference to navigate the robot to explore unknown habitats and collect information-rich data. After that, I will discuss a cooperative stochastic planning framework for multi-robot systems. This coordination method takes advantage of a classic optimal assignment problem, and re-interprets the underlying matching mechanism so that it is used to decouple and approximate the original computational prohibitive stochastic problem. Experimental and simulation results on both underwater and aerial vehicles will be included in this talk. *Biographie:* Dr. Lantao Liu is a Postdoctoral Scholar - Research Associate in the Department of Computer Science at the University of Southern California. Before his research experience at USC, Dr. Liu worked as a Postdoctoral Fellow in the Robotics Institute at Carnegie Mellon University. He received a Ph.D. from the Department of Computer Science and Engineering at Texas A&M University in 2013, and a Bachelor degree from the Department of Automatic Control at Beijing Institute of Technology in 2007. Dr. Liu's research interests include planning, decision-making, and applied machine learning methods for autonomous robotic systems, as well as distributed or decentralized coordination approaches for multi-robot or swarm systems.

1 0

Fwd: IMPORTANT - Séminaire de recrutement du Dr Konrad Kording au Département de neurosciences
by Yoshua Bengio 09 Jan '17

09 Jan '17

All those with even a mild interest in neuroscience should probably come to hear Konrad Kording. He is applying for an IVADO position which would be in the neuroscience department and would probably become a MILA faculty. Thursday at noon at room 1120 of the nearby Pavillon Desmarais. -- Yoshua ---------- Message transféré ---------- De : Crete Nancy <nancy.crete(a)umontreal.ca> Date : 9 janvier 2017 à 12:09 Objet : IMPORTANT - Séminaire de recrutement du Dr Konrad Kording au Département de neurosciences À : Bonjour, Les personnes qui souhaitent rencontrer Dr Kording doivent communiquer avec moi. *Nancy Crête* Technicienne en coordination de travail de bureau *Département de neurosciences - Direction* [image: cid:image005.jpg@01D09E00.A535F5F0] <http://medecine.umontreal.ca/recherche/les-chercheurs-de-la-faculte/campagn…> Faculté de médecine – Université de Montréal Pav. Paul-G.-Desmarais, bureau 1111 C.P. 6128, Succ. Centre-ville Montréal QC H3C 3J7 *Tél. : (514) 343-6047 <(514)%20343-6047>* Télécopieur : (514) 343-7972

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

Lisa_seminaires Janvier 2017