Auteur : plourde_mi

31 mars – Séminaire DIC-ISC-CRIA : From Decision Support Systems to Recommender Systems

Par : Pascale Zaraté

Résumé :
Decision Support Systems (DSSs) emerged in the early 1970s. We can find in the literature several definitions of these systems and a large amount of work has been done. During all these years of research the architecture of these systems has evolved, and we assisted to the introduction of Artificial Intelligence components giving to these systems the name of Recommender systems. Recommender systems are now very used in daily life platforms, i.e., Amazon etc. The way how Recommender systems are developed and implemented, considering users’ preferences, will be presented in this talk.

Bio :
Pascale Zaraté is a Professor at Toulouse 1 Capitole University. She conducts her research at the IRIT laboratory (https://www.irit.fr/~Pascale.Zarate/ ). She holds a Ph.D. in Computer Sciences / Decision Support from the LAMSADE laboratory at the Paris Dauphine University, Paris (1991). Pascale Zaraté’s current research interests include Decision Support Systems, Group Decision Support Systems, Recommender systems. She published several manuscripts: 3 books, edited 6 books, edited 18 special issues in several international journals, 11 proceedings of international conferences, 30 papers in international journals, 2 papers in national journals, 7 chapters in collective books, 52 papers in international conferences.

Jeudi 31 mars, 10 h 30. 

Lien Zoom de la rencontre : https://uqam.zoom.us/j/85407268175

Connectez-vous 10 à 15 minutes à l’avance, et donnez votre nom complet dès votre entrée pour nous faciliter la tâche de vous admettre au séminaire.

3 mars – Séminaire DIC-ISC-CRIA : Biomedical Informatics and the Science of Cognition

Par : Vimla Patel

Résumé :

My laboratory’s multidisciplinary research on medical cognition has shown the considerable importance of cognitive factors that determine how health professionals understand information, solve problems, use decision-support tools, and make decisions. These investigations into the process of diagnostic reasoning have made contributions to the design of today’s clinical AI systems. My talk will elaborate current cognitive-related research in biomedical informatics, with a focus on clinical aspects, building on earlier investigations to elucidate key lessons and challenges for the development of usable, useful, and safe decision-support systems that can augment human intelligence in the clinical world.

Bio :

Vimla L Patel is a Senior Research Scientist and Director of the Center for Cognitive Studies in Medicine and Public Health at the New York Academy of Medicine. A graduate of McGill University in Montreal, she was a Professor of Medicine and Psychology and Director of McGill Cognitive Science CenterHer early research related to cognitive mechanisms underlying expertise and medical decision-making. Her studies over the past two decades are on decision support technology and errors in complex clinical environments, addressing the role of cognition in biomedical informatics (human-technology interaction, cognitive design, distributed cognition, and team decision making) for a safer clinical workplace.

Jeudi 3 mars, 10 h 30. 

Lien Zoom de la rencontre : https://uqam.zoom.us/j/85407268175 

Connectez-vous 10 à 15 minutes à l’avance, et donnez votre nom complet dès votre entrée pour nous faciliter la tâche de vous admettre au séminaire.

17 février – Séminaire DIC-ISC-CRIA : Conscious processing, inductive biases and generalization in deep learning

Par : Yoshua Bengio

Résumé :

Humans are very good at “out-of-distribution” generalization (compared to current AI systems). It would be useful to determine the inductive biases they exploit and translate them into machine-language architectures, training frameworks and experiments. I will discuss several of these hypothesized inductive biases. Many exploit notions in causality and connect abstractions in representation learning (perception and interpretation) with reinforcement learning (abstract actions). Systematic generalizations may arise from efficient factorization of knowledge into recomposable pieces. This is partly related to symbolic AI (aas seen in the errors and limitations of reasoning in humans, as well as in our ability to learn to do this at scale, with distributed representations and efficient search). Sparsity of the causal graph and locality of interventions -- observable in the structure of sentences -- may reduce the computational complexity of both inference (including planning) and learning. This may be why evolution incorporated this as "consciousness.” I will also suggest some open research questions to stimulate further research and collaborations.


Bio :

Yoshua Bengio est professeur titulaire à l’Université de Montréal, fondateur et directeur scientifique de Mila – Institut québécois d’IA, et codirige le programme Apprentissage automatique, apprentissage biologique de CIFAR en tant que Senior Fellow. Il occupe également la fonction de directeur scientifique d’IVADO.

Jeudi 17 février, 10 h 30. 

Lien zoom : ​ https://uqam.zoom.us/j/85407268175

Connectez-vous 10 à 15 minutes à l’avance, et donnez votre nom complet dès votre entrée pour nous faciliter la tâche de vous admettre au séminaire.


10 février – Séminaire DIC-ISC-CRIA : AI and inclusion - How about African NLP projects ?

Par : Dr Houssein Ahmed Assoweh

Résumé :

AI has nowadays become an indispensable lever for economic and technological development and especially for reducing the digital divide in developing countries. However, Africa lags far behind in this area due to the weakness of its infrastructure and the lack of data to develop applications using machine learning. This lack of data is especially penalizing in the field of languages, since with more than 2000 languages spoken in Africa, few have the linguistic resources; many African languages risk disappearing in a few decades if nothing is done about it by the NLP and AI communities. Thus, the development of linguistic data remains essential to initiate cutting-edge research in AI such as neural machine translation, automatic speech recognition, language understanding, etc.

This presentation is about our current research, which concerns the automatic processing of African languages in general and the Somali language in particular, which is an Afro-Asiatic language spoken by approximately 23 M speakers living mainly in 4 countries of the horn of Africa (Somalia, Ethiopia, Kenya, Djibouti). We will discuss the different methods used to build bilingual or parallel corpora and their utility for the construction of neural machine translation systems from and to Somali, with English or French as the target or source language.

Bio :

Dr Houssein Ahmed Assoweh is currently a professor in computer science and director of the mathematics and digital research center at the University of Djibouti. He obtained a doctorate in computer science in 2019 from the University of Grenoble-Alpes, France. 

His postdoctoral studies were carried out within the NLP group of CRIA-UQAM, with the PCBF funding. His research lies at the intersection of Artificial Intelligence (AI) and Data Science (DS), and focuses on Natural Language Processing (NLP), Computational Linguistics and Machine Learning, especially for the development of artificial intelligence applications and tools for African languages among other under-resourced languages.

Jeudi 10 février, 10 h 30. 

Lien zoom :​ : ​ https://uqam.zoom.us/j/85407268175

Connectez-vous 10 à 15 minutes à l’avance, et donnez votre nom complet dès votre entrée pour nous faciliter la tâche de vous admettre au séminaire.

27 janvier – Séminaire DIC-ISC-CRIA : Machine learning for stochastic optimization

Par : Janosh Ortmann

Résumé :

In this talk, I will discuss several applications of machine learning in stochastic optimization, transport planning and humanitarian and healthcare logistics.

In stochastic programming, scenarios are used to approximate the distributions of the unknown parameters and formulate and solve multi-stage stochastic optimization models. However, optimizing with respect to each scenario is computationally costly and it is often difficult to see how a change in assumptions impacts the solution.
By considering the scenarios as data themselves and then applying unsupervised clustering methods to this data set we can obtain new insights into the underlying optimization problem. I will show how this leads to new upper and lower bounds, but also to deeper insights in applications such as humanitarian logistics and the planning of healthcare distribution networks.

If time permits, I will also discuss how reinforcement learning can allow decision makers obtain new approximatively optimal solutions.

Bio :

Janosh Ortmann Ortmann is an associate professor in Data Science and Business Intelligence at UQAM and a member of the Centre de Recherches Mathématiques (CRM) and the Group for Research in Decision Analysis (GERAD). He holds a PhD in mathematics from Warwick University and has completed postdoctoral fellowships at the University of Toronto, Université de Montréal and Concordia University.

His main research interest lies in the analysis of decision making under uncertainty, particularly using techniques from machine learning, probability theory and operations research. Currently, he is working on applications such as the design of transport networks, humanitarian logistics and personalized medicine.

Jeudi 27 janvier, 10 h 30. 

Lien zoom :​ https://uqam.zoom.us/j/85407268175

Connectez-vous 10 à 15 minutes à l’avance, et donnez votre nom complet dès votre entrée pour nous faciliter la tâche de vous admettre au séminaire.

13 janvier – Séminaire DIC-ISC-CRIA : Building Robotic intelligence that learns and accounts for human users

Par : Greg Dudek

Résumé :
In this talk I will present work we have done on developing intelligent systems that learn to operate in complex environments in response to human instruction and perceptual feedback.  In particular, I will discuss some of the ways we have exploited machine learning such as reinforcement learning to refine the performance of the both to learn high-performance navigation strategies, gaits, and a notion of curiosity-driven exploration.  I will discuss how we allow the vehicle to track human and other targets in the ocean using visual target recognition, how we have addressed walking and swimming, and how the vehicle can model human level of trust (satisfaction) in attempt to personalize it's behavior.  If time permits, I will also discuss an approach to efficient learning of human behavioural preferences and proxemics to develop robotics that are suited to operating for and with people.

Bio :
Greg Dudek is the James McGill Chair in Computer Science at McGill University, Director of the NSERC Canadian Robotics Network (NCFRN) and Vice President for Research and head of the Artificial intelligence at Samsung AI Center in Montreal. His research is on sensor-based robotics using computer vision and machine learning, as well as decision-making under uncertainty.

Jeudi 13 janvier, 10 h 30. 

Lien zoom :  https://uqam.zoom.us/j/85407268175. Connectez-vous 10 à 15 minutes à l’avance, et donnez votre nom complet dès votre entrée pour nous faciliter la tâche de vous admettre au séminaire.

Institut des sciences cognitives

Fondé en 2003, l'Institut des sciences cognitives de l'UQAM vise à favoriser la recherche et le développement de compétences dans le domaine des sciences cognitives, à en partager les connaissances, à faciliter les échanges interdisciplinaires et à animer sa communauté.

Suivez-nous

Coordonnées

Institut des sciences cognitives isc@uqam.ca
Local A-3741
400, rue Sainte-Catherine Est
Montréal (Québec) H2L 2C5