Janvier 2018 - Lisa_seminaires - mailman.iro.umontreal.ca

[Tea Talk] William Hamilton (Stanford) Fri Feb 2, 10:30AM, AA1360
by Michael Noukhovitch 01 Fév '18

01 Fév '18

*Note:* This is the same as the DIRO colloqium email, I'm just making sure that everyone on the mailing lists gets it! Sorry for the late email! This week we have a PhD Candidate from Stanford, *William Hamilton,* giving a talk on *Friday Feb 2* at* 10:30AM* in room *AA1360*. Make sure to chart your way to this tea talk and invite all your friends! Michael *KEYWORDS *ML on Graphs, Massive Structured Datasets, Social Science *TITLE* Machine Learning for Computational Social Science *ABSTRACT* The combination of machine learning and massive social datasets has the potential to revolutionize our ability to predict and understand human behavior. However, machine learning on large social datasets is difficult because these datasets tend to be noisy, dynamic, and involve graph-structured relationships (e.g., social networks between users)—while traditional machine learning tools are largely designed for static datasets comprised of simple Euclidean vectors or grids. In this talk, I will describe new methods that I have developed for machine learning on massive, graph-structured social datasets. The technical focus of the talk will be on techniques for graph embedding, i.e., representation learning on graph-structured data. In the first part, I will describe how I have used graph embedding techniques to enable diverse social applications—from modeling cultural change to predicting conflict between online communities. In the second part, I will describe a new graph embedding framework, called GraphSAGE, that can scale to datasets that are orders of magnitude larger than previous approaches and that has been deployed at the website Pinterest to power a recommender system serving over 200 million users. I will close the talk with a general outlook on social AI technologies, including future technical directions and important ethical considerations. *BIO* William (Will) Hamilton is a PhD Candidate in Computer Science at Stanford University, working jointly in the NLP and SNAP groups. He is co-advised by Dan Jurafsky and Jure Leskovec, and his interests lie at the intersection of machine learning, network science, natural language processing, and computational social science. Will's research is supported by the SAP Stanford Graduate Fellowship and an NSERC PGS-D Grant. Prior to coming to Stanford, Will completed a BSc and MSc at McGill University, where he studied machine learning under the supervision of Joelle Pineau and was the 2014 recipient of the Canadian AI Master's Thesis Award.

1 0

Fwd: Colloque du DIRO, vendredi le 2 février, 10:30, AA1360 Conférencier: William Hamilton
by Yoshua Bengio 01 Fév '18

01 Fév '18

Job talk for ML position. Candidate has a Stanford PhD. Please attend and give Simon L-J and I feedback. More candidates are coming over the next few weeks. ---------- Forwarded message ---------- From: "Pierre Poulin" <poulin(a)iro.umontreal.ca> Date: Jan 31, 2018 11:39 Subject: Colloque du DIRO, vendredi le 2 février, 10:30, AA1360 Conférencier: William Hamilton To: <seminaires(a)iro.umontreal.ca> Cc: Machine Learning for Computational Social Science par William Hamilton PhD candidate, Stanford University Vendredi 2 février, 10:30 à 11:30, salle *** 1360 ***, pavillon André-Aisenstadt Université de Montréal, 2920 chemin de la Tour Résumé: The combination of machine learning and massive social datasets has the potential to revolutionize our ability to predict and understand human behavior. However, machine learning on large social datasets is difficult because these datasets tend to be noisy, dynamic, and involve graph-structured relationships (e.g., social networks between users)—while traditional machine learning tools are largely designed for static datasets comprised of simple Euclidean vectors or grids. In this talk, I will describe new methods that I have developed for machine learning on massive, graph-structured social datasets. The technical focus of the talk will be on techniques for graph embedding, i.e., representation learning on graph-structured data. In the first part, I will describe how I have used graph embedding techniques to enable diverse social applications—from modeling cultural change to predicting conflict between online communities. In the second part, I will describe a new graph embedding framework, called GraphSAGE, that can scale to datasets that are orders of magnitude larger than previous approaches and that has been deployed at the website Pinterest to power a recommender system serving over 200 million users. I will close the talk with a general outlook on social AI technologies, including future technical directions and important ethical considerations. Bio: William (Will) Hamilton is a PhD Candidate in Computer Science at Stanford University, working jointly in the NLP and SNAP groups. He is co-advised by Dan Jurafsky and Jure Leskovec, and his interests lie at the intersection of machine learning, network science, natural language processing, and computational social science. Will's research is supported by the SAP Stanford Graduate Fellowship and an NSERC PGS-D Grant. Prior to coming to Stanford, Will completed a BSc and MSc at McGill University, where he studied machine learning under the supervision of Joelle Pineau and was the 2014 recipient of the Canadian AI Master's Thesis Award. Note: La présentation sera donnée en anglais. Pierre -- Pierre Poulin (poulin(a)iro.umontreal.ca) Directeur Dept. I.R.O., Universite de Montreal (normal mail) (courier) C.P. 6128, Succ. Centre-Ville 2920, chemin de la Tour, room 2163/2194 Montreal (Quebec) Canada, H3C 3J7 Montreal (Quebec) Canada, H3T 1J4 (514) 343-6780, (514) 343-5834 (fax) http://www.iro.umontreal.ca/~poulin

2 1

Fwd: Follow-Up on Welcome Gift from Hive!
by Yoshua Bengio 29 Jan '18

29 Jan '18

If someone is interested in labeling data, this company wrote to me with a 50$ free trial. I've no idea how good their services are. ---------- Forwarded message ---------- From: David Kloth <david(a)thehive.ai> Date: 2018-01-25 16:43 GMT-05:00 Subject: Follow-Up on Welcome Gift from Hive! To: bengioy(a)iro.umontreal.ca Dear Professor Bengio, I’m writing to follow up on the welcome kit we sent in the mail. I hope you’ve had a chance to look through it and share some of its contents with your favorite students and colleagues. The kit included your unique promo code (KEXOeePk) for $50 to try out our data labeling platform, Hive Data ( thehive.ai/academicpromo <https://app.salesforceiq.com/r?target=5a6a4f90e4b058f11924d8f8&t=AFwhZf1BYe…>), and we are also currently giving away Amazon gift cards for new referrals to Hive! One of our goals is to provide clean, open-source datasets for the entire ML community in an effort to decrease the barriers for productive AI research. I’d love to jump on a quick call to hear your thoughts on potential use cases and also answer any questions you might have on Hive. Here’s my calendar link <https://app.salesforceiq.com/r?target=5a6a4f90e4b058f11924d8f7&t=AFwhZf1BYe…> to schedule a time at your convenience. Looking forward to helping you advance your research! Sincerely, *David Kloth* Business Development thehive.ai <https://app.salesforceiq.com/r?target=5a6a4f90e4b058f11924d8f9&t=AFwhZf1BYe…> ‌

1 0

[Tea Talk] Bart Van Parys (MIT) Fri Jan 26 1:30PM AA1360
by Michael Noukhovitch 27 Jan '18

27 Jan '18

Hi all! We will have a researcher from MIT, *Bart Van Parys*, giving a talk on *Friday Jan 26th at 1:30PM* in room *AA1360*. Note the later time, *1:30!* This is a seminar is organized in collaboration with the Canada Research Chair in Decision Making under Uncertainty at GERAD! So this is ideal for all the optimization fans out there! Based on my experience, this will be a great talk and coming to it should be an easy decision! Michael *KEYWORDS *Optimization, Sparse Regression, Generalization *TITLE *Modern Optimization for Sparse Learning and Robust Analytics *ABSTRACT* We discuss the tremendous potential of integer optimization methods for learning predictive models from high-dimensional data via exact sparse regression. We show that novel integer formulations can solve exact sparse regression problems of sizes counting p=100,000s covariates for n=10,000s of samples. That is, two orders of magnitude more than current state of the art methods. We also indicate that robust optimization methods can help practitioners make data-driven decisions which are safeguarded against over-calibration to one particular data set. We claim that robust optimization methods have an enormous untapped potential when making subsequent decisions based on data. *BIO* Bart Van Parys is currently a postdoctoral researcher working with Prof. Dimitris Bertsimas at the MIT Sloan School of Management. His research interests are situated on the interface between optimization and machine learning. In 2015 he obtained his Ph.D. in control theory at the Swiss Federal Institute of Technology (ETH) in Zurich under the supervision of Prof. Manfred Morari. He received his M.E. from the University of Leuven in 2011.

1 1

Fwd: Request to disseminate RIKEN BSI Summer Program 2018
by Yoshua Bengio 26 Jan '18

26 Jan '18

---------- Forwarded message ---------- From: Kiyomi Ohno <kiyomi.ohno(a)riken.jp> Date: 2018-01-26 2:18 GMT-05:00 Subject: Request to disseminate RIKEN BSI Summer Program 2018 To: yoshua.bengio(a)umontreal.ca Dear Dr. Bengio, It is our pleasure to announce that we will have BSI Summer Program 2018, celebrating its 20th anniversary. I am writing this email to request you, one of our former prestigious lecturers, to disseminate our program to prospective candidates. I attach a flier that you can use when you forward the Summer Program information. Summer Program 2018 “*Untangling the Mysterious Brain* *– Cognition, Memory and Consciousness –* " The program will give you a chance to reflect on what our inner, subjective experiences mean, and how we can study them with scientific tools. Dates: Lecture Course: July 9-13 Internship Course: June 13-Aug.8 Place: Ohkouch Hall (lecture course), RIKEN, Saitama, Japan Application deadline: February 28, 2018 Website: http://www.brain.riken.jp/en/summer/next.html The lecturers: Michael Hausser, University College London Akiko Hayashi-Takagi, Gunma University Ryota Kanai, Araya Inc. Kiyoto Kasai, The University of Tokyo Fusao Kato, Jikei University School of Medicine Wu Li, Beijing Normal University Eve Marder, Brandeis University Reiko Mazuka, RIKEN BSI Thomas J. McHugh, RIKEN BSI Taro Toyoizumi, RIKEN BSI Inna Slutsky, Tel Aviv University Emre Yaksi, Kavli Institute, NTNU Kiyomi Ohno RIKEN Brain Science Institute Brain Science Promotion Division

1 0

[Tea Talk] Timothy O'Donnell (McGill) Fri Jan 19 10:30AM AA1360
by Michael Noukhovitch 19 Jan '18

19 Jan '18

*Note: *For the new year, we're moving to a different room, *AA1360* on the first floor! Exciting! For the first tea talk of the year we are going to have *Timothy O'Donnell*, professor at *McGill,* giving a talk on *Friday Jan 19th* at* 10:30AM* in room *AA1360*. This should be a real gronk of talk, and I'd encourage everyone to attend to it! Michael *KEYWORDS *Computational Linguistics, Language Learning, Linguistic Structure *TITLE* Computation, Storage, and Generalization in Language *ABSTRACT* A celebrated property of natural language is creativity, the ability to combine stored units to derive new expressions. This feature is found across multiple levels of linguistic structure: Sounds can be combined to form novel roots, suffixes, and affixes, which can be combined to form novel words, which can be combined to form novel sentences. However, each human language is characterized by its own inventory of units at each level and its own constraints on their combination. A syllable like "derp" sounds like a better morpheme of English than a syllable like "denp." The existence of the words "warmth" and "truth" do not imply the possibility of "coolth," but "warmness," "trueness," and "coolness" are all grammatical. And speakers of English will naturally drop the phrase "on the counter" from "John made dinner on the counter", but not from "John put dinner on the counter". How do learners acquire the inventory of units and constraints at each level of linguistic structure that are particular to their own language? I will discuss a theoretical framework designed to address this question and a number of specific applications. The approach is based on the idea that this problem can be approached by formulating a variety of linguistic learning problems as problems of program induction and then optimizing a tradeoff between a pressure to store fewer, more reusable primitive units and a pressure to account for each linguistic expression in as few computational steps as possible. Although the idea behind this tradeoff is an old one, it has surprisingly deep and far-reaching consequences when applied to domain-specific models of linguistic computation. I will show how this approach can shed light on a number of problems across the linguistic system. *BIO* Tim O'Donnell is an assistant professor in the department of linguistics at McGill University. Previously he was a research scientist at MIT in the department of Brain and Cognitive Sciences. His research focuses on developing mathematical and computational models of language learning and processing. His work draws on techniques from computational linguistics, and artificial intelligence integrating ideas from theoretical linguistics and methods from experimental psychology.

1 1

Fwd: Prochain Colloque du DIRO jeudi prochain, 18 janvier
by Yoshua Bengio 12 Jan '18

12 Jan '18

Hi all, We'll have a special guest next Thursday, Peter Railton, a philosopher teaching at U. Michichigan, Ann Arbor. He will talk on Morality and AI, note that it's not in the usual tea-talk slot but during the department colloquium (3pm for the munchies, 3:30pm for the talk). I heard him talk last Fall and I was quite impressed (about how well he understand current deep-learning based AI from a philosophical standpoint). If you are interested in chatting with him during his visit, please contact Michael (in cc) and he will set up a meeting schedule. -- Yoshua ---------- Forwarded message ---------- From: Gilles Brassard <brassard(a)iro.umontreal.ca> Date: 2018-01-11 5:28 GMT-05:00 Subject: Prochain Colloque du DIRO jeudi prochain, 18 janvier To: seminaires(a)iro.umontreal.ca Cc: Yoshua Bengio <yoshua.umontreal(a)gmail.com> Bonjour tout le monde Le premier Colloque du DIRO de la session (et donc de l'année 2018) aura lieu jeudi prochain. Ne ratez pas cette chance ! :-) Conférencier : Peter Railton Affiliation : University of Michigan, Ann Arbor Endroit : Z330 Pavillon Claire-McNicoll Université de Montréal Quand : le 18 janvier 2018, 15h30 (café et biscuits à partir de 15h) Titre : Moral Learning and Artificial Intelligence (la conférence sera présentée en anglais) Résumé : Traditional approaches to moral development have emphasized the explicit teaching of norms, e.g., via parental instruction, or the acquisition of behavioral dispositions by “social learning”, e.g., via infant imitation and modeling of observed behaviors, or progression through a fixed set of developmental “stages”. But what if we understood moral learning as closer to causal learning and the development of commonsense physics? Developmental evidence suggests that infants early on begin to model their physical environment and its possibilities (Gopnik & Schulz, 2004), using observation but receiving very limited explicit instruction or external reinforcement. Similarly, there is evidence that infants early on begin learning a kind of commonsense psychology that enables them to model others’ behavior in terms of intentional states, once again, using observation but very limited explicit instruction or external reinforcement (Wellman, 2014). These internal models enable infants to interact reasonably successfully with their physical and social environment even if they are unable to articulate the causal or psychological principles involved—the knowledge underlying these capacities is therefore generalizable despite being implicit, and so is spoken of as intuitive. Internal models are not limited to causal and predictive information, however, but also appear to encode evaluative information, including evaluation of possible actions or third-party social interactions for such features as helpfulness, harm, knowledgeability, and trustworthiness (Hamlin et al., 2011; Doebel & Koenig, 2013). When combined with an implicit capacity to empathically simulate the mental states of others, these evaluative capacities can underwrite a kind of intuitive learning of commonsense morality. Such learning occurs without much explicit instruction in moral principles, yet with a capacity to generalize and with some degree of moral autonomy—so that by age 3-4, infants will resist conforming to imposed rules that involve harm or unfairness toward others (Turiel, 2002). To be genuinely intelligent, artificial systems will need to possess the kinds of intuitive knowledge involved in commonsense physics and psychology. And to be both autonomous and trustworthy, artificial systems will need to be able to evaluate situations, actions, and agents in the terms of such categories of commonsense morality as helpfulness, harm, knowledgeability, and trustworthiness. Deep learning approaches suggest how intuitive knowledge of the kind involved in predictive learning might be acquired and represented, without being “programmed in” or explicitly taught. How might further developments of these approaches make possible the acquisition of intuitive evaluative knowledge of the kind involved in commonsense epistemic or moral assessment? Biographie : Peter Railton is the G.S. Kavka Distinguished University Professor in the Department of Philosophy at the University of Michigan, Ann Arbor. His main research has been in ethics and the philosophy of science, focusing especially on questions about the nature of objectivity, value, norms, and explanation. Recently, he has also begun working in aesthetics, moral psychology, and the theory of action, and on the bearing of empirical research in these areas. Among his publications are Facts, Values, and Norms (Cambridge), a collection of some of his papers in ethics and meta-ethics, and Homo Prospectus (Oxford), a joint project with psychology and neuroscience on basic mental architecture. He has been a visiting professor at Berkeley and Princeton, and he has received fellowships from the ACLS, the NEH, and the Guggenheim Foundation. He is a former President of the American Philosophical Association (Central Division) and is a Fellow of the American Academy of Arts and Sciences. Venez nombreux ! - Gilles et Yoshua, responsables du Colloque

1 0

No [Tea Talk] this week
by Michael Noukhovitch 10 Jan '18

10 Jan '18

Happy new year and welcome back! This week's tea talk has been delayed til March so *no tea talk this week*. Instead you can come talk to me about tea for an hour, no scheduling needed! Cheers! Michael

1 0