BEGIN:VCALENDAR
VERSION:2.0
PRODID:-// - ECPv6.16.4.1//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-ORIGINAL-URL:https://aarms.math.ca
X-WR-CALDESC:Events for 
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:UTC
BEGIN:STANDARD
TZOFFSETFROM:+0000
TZOFFSETTO:+0000
TZNAME:UTC
DTSTART:20200101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=UTC:20220412T110000
DTEND;TZID=UTC:20220412T120000
DTSTAMP:20220411T164445Z
CREATED:20220411T164445Z
LAST-MODIFIED:20220411T164445Z
UID:6644-1649761200-1649764800@aarms.math.ca
SUMMARY:AARMS Scientific Machine Learning Seminar: Michael W. Dunham (Department of Earth Sciences\, Memorial University)
DESCRIPTION:Semisupervised machine learning algorithms and their application to geoscience classification problems\n\nIn recent years\, many disciplines have been challenged with trying to efficiently extract meaning\, or value\, out of large datasets. Technological advances have improved data storage capabilities as well as how data can be obtained (e.g.\, real-time data). Manually interpreting data that are exponentially growing in volume has obvious management and analysis challenges. Machine learning is a solution to these challenges. Machine learning algorithms teach computers to recognize patterns in data and assign repetitive patterns to similar categories. This process automates pattern recognition of data and allows meaningful information to be extracted in an efficient manner.\n\n\nFor many machine learning problems\, there are sufficient data to train a wide range of algorithms. Some applications\, such as image classification and speech recognition\, have large training datasets readily available. However\, in several geoscience-related problems\, labeled data are generally obtained by sampling the earth in some manner (e.g.\, drilling wells\, field sampling\, etc.)\, which is not trivial due to cost and logistical factors. As such\, many earth science-related machine learning problems have limited training data. Supervised machine learning algorithms are prone to overfitting in scarce training data situations\, but semisupervised approaches are designed for these problems because the unlabelled data are also used to inform the learning process.\n\nThree geoscience applications inherently challenged with limited training data are well log classification\, seismic classification\, and bedrock lithology mapping. I apply various semisupervised algorithms to these three geoscience problems and determine if semisupervised algorithms can perform better than supervised methods and under what conditions\, if applicable. The semisupervised methods I consider are self-training\, label propagation\, and semisupervised Gaussian mixture models. I consider several supervised methods in my work\, but the most prevalent are gradient boosting decision tree methods (e.g.\, XGBoost\, LightGBM). The results show that semisupervised methods can outperform their supervised counterparts for each of the geoscience applications\, but there are situations where this is not always the case. Nonetheless\, semisupervised methods are rarely considered for many geoscience disciplines\, which is supported by the lack of published examples in the literature. The outcomes of this work help fill this gap\, but they also help raise the awareness of semisupervised methods.\n\n\nWebex link:\n\nhttps://mun.webex.com/mun/j.php?MTID=mf0e24b554219c531763a22ffce2e82c9
URL:https://aarms.math.ca/event/aarms-scientific-machine-learning-seminar-michael-w-dunham-department-of-earth-sciences-memorial-university/
LOCATION:WebEx seminar
CATEGORIES:AARMS Scientific Machine Learning Seminar
ORGANIZER;CN="Alexander Bihlo":MAILTO:abihlo@mun.ca
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20220405T113000
DTEND;TZID=UTC:20220405T123000
DTSTAMP:20220330T113152Z
CREATED:20220330T113152Z
LAST-MODIFIED:20220330T113152Z
UID:6634-1649158200-1649161800@aarms.math.ca
SUMMARY:AARMS Scientific Machine Learning Seminar: Geoffrey McGregor (University of Northern British Columbia)
DESCRIPTION:Conservative Hamiltonian Monte Carlo\n\nMarkov Chain Monte Carlo (MCMC) methods enable us to extract meaningful statistics from complex distributions which frequently appear in parameter estimation\, Bayesian statistics\, statistical mechanics and machine learning. Similar to how flipping a coin\, or rolling a dice\, allows us to sample from the corresponding distributions underlying these processes\, MCMC methods enable us to sample from more complex distributions. The sample statistics of the sequence generated by MCMC will converge to those of the target distribution\, or “stationary distribution” provided certain acceptance and rejection criteria are satisfied. However\, as the dimensionality of the stationary distribution increases\, the acceptance rate of traditional MCMC methods inevitably diminishes and their convergence slows down substantially. This has led to recent developments in computational techniques\, such as Hamiltonian Monte Carlo (HMC) to improve the performance in convergence and acceptance rate. Specifically\, HMC proposes samples for acceptance or rejection by solving a Hamiltonian system of differential equations using volume preserving numerical methods.\n\nIn this talk\, we introduce the Conservative Hamiltonian Monte Carlo (CHMC) method\, which instead utilizes an energy preserving numerical method\, known as the Discrete Multiplier Method. We show that CHMC converges to the correct stationary distribution under appropriate conditions and provide numerical examples showcasing improvements on acceptance rates.\n\nThis is joint work with Andy Wan from the University of Northern British Columbia.\n\nWebex link:\nhttps://mun.webex.com/mun/j.php?MTID=mdff68bb6ee4a7d34ae94a2b77b2c4888
URL:https://aarms.math.ca/event/aarms-scientific-machine-learning-seminar-geoffrey-mcgregor-university-of-northern-british-columbia/
LOCATION:WebEx seminar
CATEGORIES:AARMS Scientific Machine Learning Seminar
ORGANIZER;CN="Alexander Bihlo":MAILTO:abihlo@mun.ca
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20220329T110000
DTEND;TZID=UTC:20220329T120000
DTSTAMP:20220322T143839Z
CREATED:20220322T143839Z
LAST-MODIFIED:20220322T143839Z
UID:6631-1648551600-1648555200@aarms.math.ca
SUMMARY:AARMS Scientific Machine Learning Seminar: Scott MacLachlan (Memorial)
DESCRIPTION:Optimization and Learning in the Design of Preconditioners\n\nComputer simulation algorithms are a major tool in many areas of science and industry\, particularly in areas where the behaviour of fluids or complex materials governs the physical processes of interest.  A typical core of these tools is the numerical approximation of the solution to coupled nonlinear systems of partial differential equations\, relying on nonlinear and linear solvers\, such as Newton’s method and preconditioned Krylov iterations.  Among the most effective preconditioners for these systems are multigrid and domain decomposition methods\, which use multiscale representations of the systems to be solved to achieve linear-scaling complexity for the solution of these linear systems.  These preconditioners typically rely on heuristics in their construction\, to approximate solutions to underlying combinatorial (and other) optimization problems that specify parameters and other components of the preconditioners\, based on the discrete problem to which they are being applied.  In this talk\, I will discuss the use of advanced optimization and machine learning techniques to approximately solve these optimization problems and the impact these techniques can have on advanced preconditioner design.\n\n\nWebex link:\nhttps://mun.webex.com/mun/j.php?MTID=m8e73e9072ce63a9072b87874c1ee7cae
URL:https://aarms.math.ca/event/aarms-scientific-machine-learning-seminar-scott-maclachlan-memorial/
LOCATION:WebEx seminar
CATEGORIES:AARMS Scientific Machine Learning Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20220308T110000
DTEND;TZID=UTC:20220308T120000
DTSTAMP:20220228T160748Z
CREATED:20220228T160654Z
LAST-MODIFIED:20220228T160748Z
UID:6615-1646737200-1646740800@aarms.math.ca
SUMMARY:AARMS Scientific Machine Learning Seminar: Simone Brugiapaglia (Concordia University)
DESCRIPTION:The curse of dimensionality and the blessings of sparsity and Monte Carlo sampling: From polynomial approximation to deep learning in high dimensions \nIn data science and scientific computing\, the approximation of high-dimensional functions from pointwise samples is a ubiquitous task\, which is made intrinsically difficult by the so-called curse of dimensionality. In this talk\, we will illustrate how to alleviate the curse thanks to the “blessings” of sparsity and Monte Carlo sampling.\n\nFirst\, we will consider the case of sparse polynomial approximation via compressed sensing. Focusing on the case where the target function is smooth\, but possibly highly anisotropic\, we will show how to obtain sample complexity bounds only mildly affected by the curse of dimensionality\, near-optimal accuracy guarantees\, stability to unknown errors corrupting the data\, and rigorous convergence rates of algebraic and exponential type.\n\nThen\, we will illustrate how the mathematical toolkit of sparse polynomial approximation can be employed to obtain a “practical existence theorem” for deep learning in the context of high-dimensional Hilbert-valued function approximation. This result shows not only the existence of neural networks with desirable approximation properties\, but also how to compute them via a suitable training procedure in order to achieve best-in-class performance guarantees.\n\nWe will conclude by discussing ongoing and future research directions.\n\n\nWebex link:\n\nhttps://mun.webex.com/mun/j.php?MTID=m3908ed63dfe6896d1e2421f4a3356bc9
URL:https://aarms.math.ca/event/aarms-scientific-machine-learning-seminar-simone-brugiapaglia-concordia-university/
LOCATION:WebEx seminar
CATEGORIES:AARMS Scientific Machine Learning Seminar
ORGANIZER;CN="Alexander Bihlo":MAILTO:abihlo@mun.ca
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20220301T110000
DTEND;TZID=UTC:20220301T120000
DTSTAMP:20220224T132859Z
CREATED:20220224T132859Z
LAST-MODIFIED:20220224T132859Z
UID:6613-1646132400-1646136000@aarms.math.ca
SUMMARY:AARMS Scientific Machine Learning Seminar: Stijn De Baerdemacker (UNB)
DESCRIPTION:Deep Learning Chemistry: extracting chemical features from Graph Convolutional Neural Networks\n\nThe task of quantum chemistry is to compute chemical properties of molecular compounds from the fundamental laws of quantum mechanics.  This poses a tremendous challenge as the most accurate methods also come with a perniciously scaling computational cost\, limiting systematic and meaningful calculations mostly to medium-small molecules.  This is an undesirable situation as there is a high potential for computationally driven discovery of chemical compounds. For this reason\, there have been significant advances in the design of Machine Learning (ML) algorithms to interpolate chemical observables from precomputed data\, rather than engaging into costly computations.  At present\, the ML methods of choice in (quantum) chemistry are based upon Graph Convolutional Neural Networks (GCNN)\, as molecular compounds are easily reinterpreted into graph-based structures.\n\nIn this presentation\, I will discuss recent results from the QuNB group on the analysis of the feature space of SchNet\, a popular GCNN for quantum chemistry with high predictive power.  I will show that chemical information from chemical functional groups is learned and encoded in the feature space of the GCNN\, giving a strong quality hallmark of interpretability to the GCNN ML algorithm.\n\nJoint work from the QuNB group (UNB) with Guillaume Acke (Ghent University)\n\nWebex link: https://mun.webex.com/mun/j.php?MTID=m8b04b29275dfeb0c3617be2958e577ff
URL:https://aarms.math.ca/event/aarms-scientific-machine-learning-seminar-stijn-de-baerdemacker-unb/
CATEGORIES:AARMS Scientific Machine Learning Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20220222T110000
DTEND;TZID=UTC:20220222T120000
DTSTAMP:20220214T125321Z
CREATED:20220214T125321Z
LAST-MODIFIED:20220214T125321Z
UID:6605-1645527600-1645531200@aarms.math.ca
SUMMARY:AARMS Scientific Machine Learning Seminar:  Alison Malcom (Memorial University)
DESCRIPTION:Using Normalizing Flows for Seismic Data Interpolation\nNormalizing Flows are a type of neural network that allow us to map one probability distribution into another.  The advantage of such a technique is that they allow us to relate a simple distribution\, like a Gaussian\, to a more complicated distribution that may be more difficult to estimate and sample from.  In uncertainty quantification for inverse problems\, we are trying to estimate one of these more complicated distributions\, thus normalizing flows can help to speed up this process and improve our ability to use and analyze our results.  This will be an applied talk\, giving an introduction to normalizing flows\, discussing how we chose the particular machine learning method and explaining how it improves our understanding of seismic data processing and the associated uncertainties.
URL:https://aarms.math.ca/event/aarms-scientific-machine-learning-seminar-alison-malcom-memorial-university/
LOCATION:WebEx seminar
CATEGORIES:AARMS Scientific Machine Learning Seminar
ORGANIZER;CN="Alexander Bihlo":MAILTO:abihlo@mun.ca
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20211207T110000
DTEND;TZID=UTC:20211207T120000
DTSTAMP:20211126T123922Z
CREATED:20211126T123922Z
LAST-MODIFIED:20211126T123922Z
UID:6540-1638874800-1638878400@aarms.math.ca
SUMMARY:AARMS Scientific Machine Learning Seminar: Peter Dueben (ECMWF)
DESCRIPTION:This talk provides an overview on the machine learning efforts at the European Centre for Medium-Range Weather Forecasts (ECMWF)\, and outlines how machine learning\, and in particular deep learning\, could help to improve weather predictions in the coming years. The talk will name challenges for the use of machine learning and suggest developments (research/software/hardware) that should enable the community of Earth system modelling to make quick progress. \nWebex information: \nLink: https://mun.webex.com/mun/j.php?MTID=m32a22bd61d05707cefe973ca5bf6ad69 \nMeeting no: 2631 545 3588 \nPassword: t32XHrsMqu6
URL:https://aarms.math.ca/event/aarms-scientific-machine-learning-seminar-peter-dueben-ecmwf/
LOCATION:WebEx seminar
CATEGORIES:AARMS Scientific Machine Learning Seminar
ORGANIZER;CN="Alexander Bihlo":MAILTO:abihlo@mun.ca
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20211123T110000
DTEND;TZID=UTC:20211123T120000
DTSTAMP:20211117T110237Z
CREATED:20211112T192613Z
LAST-MODIFIED:20211117T110237Z
UID:6497-1637665200-1637668800@aarms.math.ca
SUMMARY:AARMS Scientific Machine Learning Seminar: Ben Adcock (Simon Fraser University)
DESCRIPTION:Title: Tackling the curse: polynomial and deep neural network methods for function approximation in high dimensions \nAbstract: Many problems in computational science and engineering require the accurate approximation of a target function from data. This problem is rendered challenging by the high-dimensionality of the function\, the expense of generating function samples\, the presence of noise in the measurements\, and the fact that the target function may take values in a function space. Developing techniques that tackle these challenges without succumbing to the famous “curse of dimensionality” has been a long-standing problem. In the first part of this talk I will give a brief survey of a decade’s worth of progress on high-dimensional function approximation via sparse polynomial expansions. I will show how the proper use of nonlinear approximation theory and compressed sensing techniques leads to algorithms for high-dimensional approximation which\, unlike other approaches\, possess provably near-optimal error bounds and moderate sample complexities. In particular\, these techniques mitigate the curse of dimensionality to a substantial degree. The second part of the talk will be devoted to emerging approaches based on deep neural networks and deep learning. Such tools are beginning to garner substantial attention in the scientific computing community. Nonetheless\, I will present evidence of a key gap between current theory and practice. I will then discuss recent results showing that there exist deep neural networks that match the performance of best-in-class schemes\, and furthermore\, these can indeed be trained through realizable procedures. This highlights the potential of deep neural networks\, and sheds light on achieving robust\, reliable and overall improved practical performance. \nThis talk is based on joint work with Anyi Bao\, Simone Brugiapaglia\, Juan M. Cardenas\, Nick Dexter\, Sebastian Moraga\, Yi Sui and Clayton G. Webster. \nMeeting link:\nhttps://mun.webex.com/mun/j.php?MTID=m38cc6ce370de2b1c234e33feac5bb93a \nMeeting number: 2630 760 7617\nPassword: cJwPcrHM425
URL:https://aarms.math.ca/event/aarms-scientific-machine-learning-seminar-ben-adcock-sfu/
LOCATION:WebEx seminar
CATEGORIES:AARMS Scientific Machine Learning Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20211109T110000
DTEND;TZID=UTC:20211109T120000
DTSTAMP:20211113T004815Z
CREATED:20211102T153203Z
LAST-MODIFIED:20211113T004815Z
UID:6474-1636455600-1636459200@aarms.math.ca
SUMMARY:AARMS Scientific Machine Learning Seminar: Hamid Usefi (MUN)
DESCRIPTION:Multicollinearity\, singular vectors\, and dimensionality reduction for high-dimensional datasets\nSingle nucleotide polymorphisms (SNPs) as building blocks of our DNA\, can determine the variations between people. It is believed that SNPs in genes that regulate DNA mismatch repair\, cell cycle regulation\, metabolism and immunity are associated with genetic susceptibility to cancer.  So\, SNPs are potential diagnostic and therapeutic biomarkers in many cancer types. This in part has prompted the rapid advancements in DNA sequencing  which makes it  possible both in terms of cost and time to genetically sequence a single suspect tissue. The number of SNPs in a disease dataset varies from tens of thousands to several million. So\, one of the bottlenecks of working with these genome datasets is their large-scale size that makes it difficult to render the data for meaningful analysis.  Furthermore\, in most diseases\, there are at most a couple of hundred SNPs associated with the disease. So\, simply put we would be looking for a needle in a haystack. \nMachine learning algorithms are gaining increasing attention and believe to have great potential in answering many questions in this respect. A common problem in machine learning and pattern recognition is the process of identifying the most relevant features\, specifically in dealing with high-dimensional datasets in bioinformatics. In this talk\, I will discuss some of our recent work on a new feature selection method\, called Singular-Vectors Feature Selection (SVFS). Part of this work is joint with my recently graduated PhD student Majid Afshar. It is stemmed from identifying linearly dependent columns of a matrix A. This problem can also be viewed as multicollinearity and subset selection in statistical modelling and arises in many contexts\, including regression\, ecology\, and machine learning. \nLet D = [A | b] be a labeled dataset\, where b is the class label and features (attributes or SNPs) are columns of matrix A; rows of A can be viewed as samples. We show with examples as well as a sketch of proof that the projector matrix P_A onto the null space of A can be used to partition the columns of A into clusters so that columns in a cluster correlate only with the columns in the same cluster. In the first step\, SVFS uses the projector P_D to find the cluster that contains b. We reduce the size of A by discarding features in the other clusters as irrelevant features. In the next step\, SVFS uses the P_A of reduced A to partition the remaining features into clusters and choose the most important features from each cluster. I will discuss the  performance of SVFS on genomic datasets compared to the state-of-the-art feature selection methods. \n[ recording ] \n\nMeeting link:  https://mun.webex.com/mun/j.php?MTID=m855b1e73549cf668f5b57190a7ef3eae\nMeeting number:   2633 483 6250\nMeeting password:  W3FisMnJa86\n\n\n 
URL:https://aarms.math.ca/event/aarms-scientific-machine-learning-seminar-hamid-usefi-mun/
LOCATION:WebEx seminar
CATEGORIES:AARMS Scientific Machine Learning Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20211026T110000
DTEND;TZID=UTC:20211026T120000
DTSTAMP:20211117T110211Z
CREATED:20211019T225913Z
LAST-MODIFIED:20211117T110211Z
UID:6371-1635246000-1635249600@aarms.math.ca
SUMMARY:AARMS Scientific Machine Learning Seminar: Nicholas Touikan (University of New Brunswick)
DESCRIPTION:Group equivariant neural networks seen by a mathematician\nArtificial neural networks (ANNs) are incredibly successful at performing certain machine learning tasks\, such as classification. In applications such as computer vision or quantum chemistry\, we will often seek machine learning algorithms that can handle inputs that are transformed. For example\, a cat detector should be able to detect a rotated cat. \nGroup theory provides the natural formalization of what we mean by transformations and group equivariance is the property we seek in artificial neural networks (ANN) and there is currently a flurry of research activity in group equivariant neural networks. In this talk\, I will present the M.Sc. work of my former student Max Hennick\, which gives a characterization of (approximate) G-equivariant linear mappings. What is most striking is how effective a bit of functional analysis and algebra can be at answering this question. \nI will provide as many examples as possible and conclude with some hopefully interesting questions. \n[ recording ] \nThe AARMS Scientific Machine Learning Seminar takes virtually via WebEx.  If you would like to attend\, please email the organizers for connection details.
URL:https://aarms.math.ca/event/aarms-scientific-machine-learning-seminar-nicholas-touikan-unb/
LOCATION:WebEx seminar
CATEGORIES:AARMS Scientific Machine Learning Seminar
END:VEVENT
END:VCALENDAR