BEGIN:VCALENDAR VERSION:2.0 PRODID:-//Drupal iCal API//EN X-WR-CALNAME:Events items teaser X-WR-TIMEZONE:America/Toronto BEGIN:VTIMEZONE TZID:America/Toronto X-LIC-LOCATION:America/Toronto BEGIN:DAYLIGHT TZNAME:EDT TZOFFSETFROM:-0500 TZOFFSETTO:-0400 DTSTART:20200308T070000 END:DAYLIGHT BEGIN:DAYLIGHT TZNAME:EDT TZOFFSETFROM:-0500 TZOFFSETTO:-0400 DTSTART:20190310T070000 END:DAYLIGHT BEGIN:DAYLIGHT TZNAME:EDT TZOFFSETFROM:-0500 TZOFFSETTO:-0400 DTSTART:20180311T070000 END:DAYLIGHT BEGIN:STANDARD TZNAME:EST TZOFFSETFROM:-0400 TZOFFSETTO:-0500 DTSTART:20191103T060000 END:STANDARD BEGIN:STANDARD TZNAME:EST TZOFFSETFROM:-0400 TZOFFSETTO:-0500 DTSTART:20181104T060000 END:STANDARD BEGIN:STANDARD TZNAME:EST TZOFFSETFROM:-0400 TZOFFSETTO:-0500 DTSTART:20171105T060000 END:STANDARD END:VTIMEZONE BEGIN:VEVENT UID:69f3c02ec46ea DTSTART;TZID=America/Toronto:20201005T093000 SEQUENCE:0 TRANSP:TRANSPARENT DTEND;TZID=America/Toronto:20201005T093000 URL:https://uwaterloo.ca/artificial-intelligence-group/events/phd-defence-c omputational-mechanisms-language-understanding LOCATION:Online 200 University Avenue West Waterloo ON N2L 3G1 Canada SUMMARY:PhD Defence: Computational Mechanisms of Language Understanding and \nUse in the Brain and Behaviour CLASS:PUBLIC DESCRIPTION:PLEASE NOTE: THIS PHD DEFENCE WILL BE GIVEN ONLINE.\n\nIVANA KA JIĆ\, PHD CANDIDATE\n_David R. Cheriton School of Computer Science_\n\nS UPERVISOR: Professor Chris Eliasmith DTSTAMP:20260430T204846Z END:VEVENT BEGIN:VEVENT UID:69f3c02ec6398 DTSTART;TZID=America/Toronto:20191127T100000 SEQUENCE:0 TRANSP:TRANSPARENT DTEND;TZID=America/Toronto:20191127T100000 URL:https://uwaterloo.ca/artificial-intelligence-group/events/phd-defence-e motion-aware-and-human-autonomous-agents LOCATION:DC - William G. Davis Computer Research Centre 200 University Aven ue West 2306C Waterloo ON N2L 3G1 Canada SUMMARY:PhD Defence: Emotion-Aware and Human-Like Autonomous Agents CLASS:PUBLIC DESCRIPTION:NABIHA ASGHAR\, PHD CANDIDATE\n_David R. Cheriton School of Com puter Science_\n\nIn human-computer interaction (HCI)\, one of the technol ogical goals is\nto build human-like artificial agents that can think\, de cide and\nbehave like humans during the interaction. A prime example is a\ ndialogue system\, where the agent should converse fluently and\ncoherentl y with a user and connect with them emotionally. Humanness\nand emotion-aw areness of interactive artificial agents have been shown\nto improve user experience and help attain application-specific goals\nmore quickly. Howev er\, achieving human-likeness in HCI systems is\ncontingent on addressing several philosophical and scientific\nchallenges. In this thesis\, I addre ss two such challenges: replicating\nthe human ability to 1) correctly per ceive and adopt emotions\, and 2)\ncommunicate effectively through languag e. DTSTAMP:20260430T204846Z END:VEVENT BEGIN:VEVENT UID:69f3c02ec7309 DTSTART;TZID=America/Toronto:20191206T140000 SEQUENCE:0 TRANSP:TRANSPARENT DTEND;TZID=America/Toronto:20191206T140000 URL:https://uwaterloo.ca/artificial-intelligence-group/events/phd-defence-l ikelihood-based-density-estimation-using-deep LOCATION:DC - William G. Davis Computer Research Centre 200 University Aven ue West 2310 Waterloo ON N2L 3G1 Canada SUMMARY:PhD Defence: Likelihood-based Density Estimation using Deep\nArchit ectures CLASS:PUBLIC DESCRIPTION:PRIYANK JAINI\, PHD CANDIDATE\n_David R. Cheriton School of Com puter Science_\n\nMultivariate density estimation is a central problem in unsupervised\nmachine learning that has been studied immensely in both sta tistics\nand machine learning. Several methods have thus been proposed for \ndensity estimation including classical techniques like histograms\,\nker nel density estimation methods\, mixture models\, and more recently\nneura l density estimation that leverages the recent advances in deep\nlearning and neural networks to tractably represent a density\nfunction. In today' s age when large amounts of data are being\ngenerated in almost every fiel d it is of paramount importance to\ndevelop density estimation methods tha t are cheap both computationally\nand in memory cost. The main contributio n of this thesis is in\nproviding a principled study of parametric density estimation methods\nusing mixture models and triangular maps for neural d ensity\nestimation.  DTSTAMP:20260430T204846Z END:VEVENT BEGIN:VEVENT UID:69f3c02ec8074 DTSTART;TZID=America/Toronto:20190529T113000 SEQUENCE:0 TRANSP:TRANSPARENT DTEND;TZID=America/Toronto:20190529T113000 URL:https://uwaterloo.ca/artificial-intelligence-group/events/phd-defence-t heoretical-foundations-efficient-clustering LOCATION:DC - William G. Davis Computer Research Centre 200 University Aven ue West 2310 Waterloo ON N2L 3G1 Canada SUMMARY:PhD Defence: Theoretical Foundations for Efficient Clustering CLASS:PUBLIC DESCRIPTION:SHRINU KUSHAGRA\, PHD CANDIDATE\n_David R. Cheriton School of C omputer Science_\n\nClustering aims to group together data instances which are similar\nwhile simultaneously separating the dissimilar instances. Th e task of\nclustering is challenging due to many factors. The most well-st udied\nis the high computational cost. The clustering task can be viewed a s\nan optimization problem where the goal is to minimize a certain cost\nf unction (like k-means cost or k-median cost). Not only are the\nminimizati on problems NP-Hard but often also NP-Hard to approximate\n(within a const ant factor). There are two other major issues in\nclustering\, namely unde r-specificity and noise-robustness.  DTSTAMP:20260430T204846Z END:VEVENT BEGIN:VEVENT UID:69f3c02ec8db4 DTSTART;TZID=America/Toronto:20190425T130000 SEQUENCE:0 TRANSP:TRANSPARENT DTEND;TZID=America/Toronto:20190425T130000 URL:https://uwaterloo.ca/artificial-intelligence-group/events/phd-defence-d ynamical-systems-spiking-neuromorphic-hardware LOCATION:DC - William G. Davis Computer Research Centre 200 University Aven ue West 2310 Waterloo ON N2L 3G1 Canada SUMMARY:PhD Defence: Dynamical Systems in Spiking Neuromorphic Hardware CLASS:PUBLIC DESCRIPTION:AARON VOELKER\, PHD CANDIDATE\n_David R. Cheriton School of Com puter Science_\n\nDynamical systems are universal computers. They can perc eive stimuli\,\nremember\, learn from feedback\, plan sequences of actions \, and\ncoordinate complex behavioural responses. The Neural Engineering\n Framework (NEF) provides a general recipe to formulate models of such\nsys tems as coupled sets of nonlinear differential equations and\ncompile them onto recurrently connected spiking neural networks —\nakin to a program ming language for spiking models of computation. The\nNengo software ecosy stem supports the NEF and compiles such models\nonto neuromorphic hardware .  DTSTAMP:20260430T204846Z END:VEVENT BEGIN:VEVENT UID:69f3c02ec98e2 DTSTART;TZID=America/Toronto:20181022T160000 SEQUENCE:0 TRANSP:TRANSPARENT DTEND;TZID=America/Toronto:20181022T160000 URL:https://uwaterloo.ca/artificial-intelligence-group/events/phd-defence-t witsong-current-events-computer-poet-and-thorny LOCATION:DC - William G. Davis Computer Research Centre 200 University Aven ue West 2310 Waterloo ON N2L 3G1 Canada SUMMARY:PhD Defence: TwitSong: A Current Events Computer Poet and the Thorn y\nProblem of Assessment CLASS:PUBLIC DESCRIPTION:Carolyn Lamb\, PhD candidate\n_David R. Cheriton School of Comp uter Science_\n\nThis thesis is driven by the question of how computers ca n generate\npoetry\, and how that poetry can be evaluated. We survey exist ing work\non computer-generated poetry and interdisciplinary work on how t o\nevaluate this type of computer-generated creative product.  DTSTAMP:20260430T204846Z END:VEVENT BEGIN:VEVENT UID:69f3c02eca3e6 DTSTART;TZID=America/Toronto:20180924T133000 SEQUENCE:0 TRANSP:TRANSPARENT DTEND;TZID=America/Toronto:20180924T133000 URL:https://uwaterloo.ca/artificial-intelligence-group/events/masters-thesi s-presentation-computational-complexity-center LOCATION:DC - William G. Davis Computer Research Centre 200 University Aven ue West 2310 Waterloo ON N2L 3G1 Canada SUMMARY:Master’s Thesis Presentation: On the Computational Complexity of\ nCenter-based Clustering CLASS:PUBLIC DESCRIPTION:NICOLE MCNABB\, MASTER’S CANDIDATE\n_David R. Cheriton School of Computer Science_\n\nClustering is the task of partitioning data so th at “similar”\npoints are grouped together and “dissimilar” ones ar e separated.\nIn general\, this is an ill-defined task. One way to make cl ustering\nwell-defined is to introduce a clustering objective to optimize. While\nmany common objectives such as k-means are known to be NP-hard\,\n heuristics output “nice” clustering solutions efficiently in\npractice . This work analyzes two avenues of theoretical research that\nattempt to explain this discrepancy. DTSTAMP:20260430T204846Z END:VEVENT BEGIN:VEVENT UID:69f3c02ecad1a DTSTART;TZID=America/Toronto:20180913T133000 SEQUENCE:0 TRANSP:TRANSPARENT DTEND;TZID=America/Toronto:20180913T133000 URL:https://uwaterloo.ca/artificial-intelligence-group/events/masters-thesi s-presentation-predicting-short-term-water LOCATION:DC - William G. Davis Computer Research Centre 200 University Aven ue West 2310 Waterloo ON N2L 3G1 Canada SUMMARY:Master’s Thesis Presentation: Predicting Short-Term Water\nConsum ption for Multi-Family Residences CLASS:PUBLIC DESCRIPTION:IRISH MEDINA\, MASTER’S CANDIDATE\n_David R. Cheriton School of Computer Science_\n\nSmart water meters have been installed across Abbo tsford\, British\nColumbia\, Canada\, to measure the water consumption of households in\nthe area. Using this water consumption data\, we develop ma chine\nlearning and deep learning models to predict daily water consumptio n\nfor existing multi-family residences. We also present a new\nmethodolog y for predicting the water consumption of new housing\ndevelopments.  DTSTAMP:20260430T204846Z END:VEVENT BEGIN:VEVENT UID:69f3c02ecb6a0 DTSTART;TZID=America/Toronto:20180925T130000 SEQUENCE:0 TRANSP:TRANSPARENT DTEND;TZID=America/Toronto:20180925T130000 URL:https://uwaterloo.ca/artificial-intelligence-group/events/phd-defence-l earning-sparse-orthogonal-wavelet-filters LOCATION:DC - William G. Davis Computer Research Centre 200 University Aven ue West 2310 Waterloo ON N2L 3G1 Canada SUMMARY:PhD Defence: Learning Sparse Orthogonal Wavelet Filters CLASS:PUBLIC DESCRIPTION:Daniel Recoskie\, PhD candidate\nDavid R. Cheriton School of Co mputer Science\n\nThe wavelet transform is a well-studied and understood a nalysis\ntechnique used in signal processing. In wavelet analysis\, signal s are\nrepresented by a sum of self-similar wavelet and scaling functions. \nTypically\, the wavelet transform makes use of a fixed set of wavelet\nf unctions that are analytically derived. We propose a method for\nlearning wavelet functions directly from data. We impose an\northogonality constrai nt on the functions so that the learned wavelets\ncan be used to perform b oth analysis and synthesis. We accomplish this\nby using gradient descent and leveraging existing automatic\ndifferentiation frameworks. Our learned wavelets are able to capture\nthe structure of the data by exploiting spa rsity. We show that the\nlearned wavelets have similar structure to tradit ional wavelets. DTSTAMP:20260430T204846Z END:VEVENT BEGIN:VEVENT UID:69f3c02ecbef3 DTSTART;TZID=America/Toronto:20180809T100000 SEQUENCE:0 TRANSP:TRANSPARENT DTEND;TZID=America/Toronto:20180809T100000 URL:https://uwaterloo.ca/artificial-intelligence-group/events/masters-thesi s-presentation-disentangled-representation LOCATION:DC - William G. Davis Computer Research Centre 200 University Aven ue West 3102 Waterloo ON N2L 3G1 Canada SUMMARY:Master’s Thesis Presentation: Disentangled Representation Learnin g\nfor Stylistic Variation in Neural Language Models CLASS:PUBLIC DESCRIPTION:Vineet John\, Master’s candidate\nDavid R. Cheriton School of Computer Science\n\nThis thesis tackles the problem of disentangling the latent style and\ncontent variables in a language modelling context. This involves\nsplitting the latent representations of documents by learning wh ich\nfeatures of a document are discriminative of its style and content\,\ nand encoding these features separately using neural network models. DTSTAMP:20260430T204846Z END:VEVENT END:VCALENDAR