BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//ZContent.net//ZapCalLib 1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:America/Vancouver
BEGIN:DAYLIGHT
DTSTART:20190310T030000
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20191103T010000
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20200308T030000
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20201101T010000
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20210314T030000
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20211107T010000
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
SUMMARY:Symposium seminar: Adaptive MCMC For Everyone from Dr. Jeffrey Rose
nthal
DTSTART:20210114T010000Z
DTEND:20210114T020000Z
UID:16
LOCATION:Zoom recording Passcode: Nzy8PB&\;A
DESCRIPTION:Dr. Jeffrey Rosenthal\nTitle: Adaptive
MCMC For Everyone\nLocation: Zoom recording Passcode: Nzy8PB&\;A\nAbstr
act: Markov chain Monte Carlo (MCMC) algorithms, such as the Metropolis Al
gorithm and the Gibbs Sampler, are an extremely useful and popular method
of approximately sampling from complicated probability distributions. Adap
tive MCMC attempts to automatically modify the algorithm while it runs, to
improve its performance on the fly. However, such adaptation often destro
ys the ergodicity properties necessary for the algorithm to be valid. In t
his talk, we first illustrate MCMC algorithms using simple graphical examp
les. We then discuss adaptive MCMC, and present examples and theorems conc
erning its ergodicity and efficiency.
DTSTAMP:20211205T081730Z
END:VEVENT
BEGIN:VEVENT
SUMMARY:Symposium seminar: The UK Biobank: Brain Imaging at the Population
Scale from Dr. Karla Miller
DTSTART:20210217T170000Z
DTEND:20210217T180000Z
UID:17
LOCATION:https://people.ok.ubc.ca/ylucet/symposium/seminars.php?range=past
DESCRIPTION:Dr. Karla
Miller\nTitle: The UK Biobank: Brain Imaging at the Population Scale\
nLocation: https://people.ok.ubc.ca/ylucet/symposium/seminars.php?range=pa
st\nAbstract: The UK Biobank is currently conducting the most ambitious im
aging study ever undertaken, scanning 100,000 subjects, with brain imaging
as a key component. In this talk, I'll overview the Biobank neuroimaging
resource, which includes six separate imaging modalities. I'll then presen
t several studies that highlight how this resource is already revolutionis
ing the neuroimaging field by enabling both discovery science and hypothes
is-driven research that previously would not have been possible.
DTSTAMP:20211205T081730Z
END:VEVENT
BEGIN:VEVENT
SUMMARY:Symposium seminar: Bio-inspired Swarm Robotics for Planar Construct
ion from Dr. Andrew Vardy
DTSTART:20210401T190000Z
DTEND:20210401T200000Z
UID:18
LOCATION:https://ubc.zoom.us/j/65764314667?pwd=bkVxSmRsNmxybzNQUHczUG5YM21K
Zz09
DESCRIPTION:Dr. Andrew Vardy\nTitle: Bio-inspired Swarm Robotics for Planar
Construction\nLocation: https://ubc.zoom.us/j/65764314667?pwd=bkVxSmRsNmx
ybzNQUHczUG5YM21KZz09\nAbstract: In this talk I will review my research gr
oup's work on swarm robotics and how we get our robots to move objects int
o desired configurations, a task we call planar construction. We are inspi
red by social insects which use pheromones to help guide the construction
of their nests. In a similar way we define scalar fields that guide the ro
bots as they gather objects together or form them into a desired shape. I
will show some of our results in simulation as well as some limited trials
on physical robots.
DTSTAMP:20211205T081730Z
END:VEVENT
BEGIN:VEVENT
SUMMARY:Symposium seminar: Using dynamical geometry tools to study algorith
ms from Dr. Scott B Lindstrom
DTSTART:20210929T230000Z
DTEND:20210930T000000Z
UID:19
LOCATION:https://ubc.zoom.us/j/67371297541?pwd=UkczdHI5dWNBK0FWTCtsaW1qRlJj
dz09
DESCRIPTION:Dr. Scott B Lindstrom\nTitle: Using dynamical ge
ometry tools to study algorithms\nLocation: https://ubc.zoom.us/j/67371297
541?pwd=UkczdHI5dWNBK0FWTCtsaW1qRlJjdz09\nAbstract: I will describe how I
used the free geometry tool Cinderella to discover and describe a class of
algorithms that may be useful for nonlinear optimization. This presentati
on will be conducted entirely within the dynamical interface, so that all
of the concepts are very visible. It should be accessible to, and the tool
s-based approach may be useful for, a wide audience. The presentation is p
rincipally based on the recent article:\n\nLindstrom, Scott B. "Computable
centering methods for spiraling algorithms and their duals, with motivati
ons from the theory of Lyapunov functions." arXiv preprint arXiv:2001.1078
4 (2020).\n
DTSTAMP:20211205T081730Z
END:VEVENT
BEGIN:VEVENT
SUMMARY:Symposium seminar: Q/A on Using dynamical geometry tools to study a
lgorithms from Dr. Scott B Lindstrom
DTSTART:20210930T000000Z
DTEND:20210930T003000Z
UID:20
LOCATION:https://ubc.zoom.us/https://ubc.zoom.us/j/65321275119?pwd=NUlwaE0z
V2thOVJ4UWlRUDJwcVp1QT09meeting/register/u5Etdu2qrTktEtXr8ka0N7sPgzrpt-byE
ScK
DESCRIPTION:Dr. Scott B Lindstrom\nTitle: Q/A on Using dynam
ical geometry tools to study algorithms\nLocation: https://ubc.zoom.us/htt
ps://ubc.zoom.us/j/65321275119?pwd=NUlwaE0zV2thOVJ4UWlRUDJwcVp1QT09meeting
/register/u5Etdu2qrTktEtXr8ka0N7sPgzrpt-byEScK\nAbstract: I will describe
how I used the free geometry tool Cinderella to discover and describe a cl
ass of algorithms that may be useful for nonlinear optimization. This pres
entation will be conducted entirely within the dynamical interface, so tha
t all of the concepts are very visible. It should be accessible to, and th
e tools-based approach may be useful for, a wide audience. The presentatio
n is principally based on the recent article:\n\nLindstrom, Scott B. "Comp
utable centering methods for spiraling algorithms and their duals, with mo
tivations from the theory of Lyapunov functions." arXiv preprint arXiv:200
1.10784 (2020).\n
DTSTAMP:20211205T081730Z
END:VEVENT
BEGIN:VEVENT
SUMMARY:Symposium seminar: Q/A from Hansi Singh
DTSTART:20211116T200000Z
DTEND:20211116T203000Z
UID:23
LOCATION:https://ubc.zoom.us/j/63425637433?pwd=cStxWWZMeHJoLzFPeGpORnhFcDFW
QT09
DESCRIPTION:Hansi Singh\nTitle:
Q/A\nLocation: https://ubc.zoom.us/j/63425637433?pwd=cStxWWZMeHJoLzFPeGpO
RnhFcDFWQT09\nAbstract: NA
DTSTAMP:20211205T081730Z
END:VEVENT
BEGIN:VEVENT
SUMMARY:Symposium seminar: Global Climate and the Ocean: Understanding Coup
led Dynamics through a Unified Framework of Energetics and Feedbacks from
Hansi Singh
DTSTART:20211116T190000Z
DTEND:20211116T200000Z
UID:27
LOCATION:https://ubc.zoom.us/j/66623530555?pwd=S1RuWlh3cWIvV0t1Z084RnNuRnk5
dz09
DESCRIPTION:Hansi Singh\nTitle:
Global Climate and the Ocean: Understanding Coupled Dynamics through a Un
ified Framework of Energetics and Feedbacks\nLocation: https://ubc.zoom.us
/j/66623530555?pwd=S1RuWlh3cWIvV0t1Z084RnNuRnk5dz09\nAbstract: The Earth's
climate arises from a complex interplay between its various components: t
he atmosphere, ocean, land surface, and ice. In this talk, I present a ser
ies of studies showing the extent to which coupled dynamics, namely intera
ctions between the ocean and atmosphere, drive the evolution of the climat
e system and its sensitivity to forcings. In these studies, I use state-of
-the-art Earth system models to show that the ocean and its dynamics impac
t climate sensitivity (how much the planet warms with CO2-doubling), preci
pitation sensitivity (how much precipitation changes per unit of warming),
and surface climate variability over a range of time scales. I conclude b
y describing how better understanding of such coupled (ocean-atmosphere) d
ynamics are likely key for designing optimal climate intervention strategi
es, which may help avert some of the irreversible impacts of planetary war
ming as human societies begin the slow transition away from fossil fuels.
DTSTAMP:20211205T081730Z
END:VEVENT
BEGIN:VEVENT
SUMMARY:Symposium seminar: MATH 590: graduate presentations from MATH 590
DTSTART:20211123T190000Z
DTEND:20211123T210000Z
UID:30
LOCATION:https://ubc.zoom.us/j/69780744256?pwd=TStzZDJlWW9TYnZtajJrSjlKUlQ1
Zz09
DESCRIPTION:MATH 590\nTitle: MATH 590: graduate presentation
s\nLocation: https://ubc.zoom.us/j/69780744256?pwd=TStzZDJlWW9TYnZtajJrSjl
KUlQ1Zz09\nAbstract: 11:00-11:30 Ziyuan Wang\nError bounds, quadratic grow
th, and linear convergence of prox-linear algorithm\n\n11:30-12:00 Alex Ia
nnantuono\nBenson's Algorithm: An Outer Approximation Algorithm for Genera
ting All Efficient Extreme Points in the Outcome Set of a Multiple Objecti
ve Linear Programming Problem\n\n12:00-12:30 Tareque Hossian\nImpact of Va
rying Community Networks on Disease Invasion\n\n12:30-13:00 Stephanie Hami
lton\nA Review of Improved filtering for the bin-packing with cardinality
constraint
DTSTAMP:20211205T081730Z
END:VEVENT
BEGIN:VEVENT
SUMMARY:Symposium seminar: MATH 590: graduate presentations from MATH 590
DTSTART:20211130T190000Z
DTEND:20211130T203000Z
UID:31
LOCATION:https://ubc.zoom.us/j/69780744256?pwd=TStzZDJlWW9TYnZtajJrSjlKUlQ1
Zz09
DESCRIPTION:MATH 590\nTitle: MATH 590: graduate presentation
s\nLocation: https://ubc.zoom.us/j/69780744256?pwd=TStzZDJlWW9TYnZtajJrSjl
KUlQ1Zz09\nAbstract: 11:00-11:30 Sarah Wyse\nThe story of structural sensi
tivity\n\n11:30-12:00 Neha Bansal\nUnderstanding Invariant Function and Me
asure for Discrete Time Markov Process via Renewal Approach\n\n12:00-12:30
Alex (Xiaoyu) Mao\nOptimization of Inf-Convolution Regularized Nonconvex
Composite Problems\n
DTSTAMP:20211205T081730Z
END:VEVENT
BEGIN:VEVENT
SUMMARY:Symposium seminar: Formulation and solution of stochastic inverse p
roblems for science and engineering models from Dr Donald Estep
DTSTART:20220215T000000Z
DTEND:20220215T010000Z
UID:44
LOCATION:https://ubc.zoom.us/j/62574960025?pwd=M0RPYTZ3SlZLaEl5b1RTR0p0alNk
Zz09
DESCRIPTION:Dr Donald Estep\nTitle
: Formulation and solution of stochastic inverse problems for science and
engineering models\nLocation: https://ubc.zoom.us/j/62574960025?pwd=M0RPYT
Z3SlZLaEl5b1RTR0p0alNkZz09 \nAbstract: Determining information about the s
tate of a complex physical system from observations of its behavior is a f
undamental problem in scientific inference and engineering design. Often,
this can be formulated as the stochastic inverse problem of determining pr
obability structures on parameters for a physics model corresponding to a
probability structure on the output of the model. We describe the formulat
ion and solution of stochastic inverse problems. Our approach yields a com
putationally tractable problem while avoiding alterations of the model lik
e regularization and ad hoc assumptions about the probability structures.
We present several examples, including a high-dimensional application to d
etermination of parameter fields in storm surge models. We describe severa
l extensions and on-going research.
DTSTAMP:20220203T014218Z
END:VEVENT
BEGIN:VEVENT
SUMMARY:Symposium seminar: High-Density Scintillating Glass Detector Use in
Proton Computed Tomography from Adam Zeiser
DTSTART:20220920T223000Z
DTEND:20220921T000000Z
UID:52
LOCATION:https://goo.gl/maps/XYQ7RgmwhowMYpUcA
DESCRIPTION:Adam Zei
ser\nTitle: High-Density Scintillating Glass Detector Use in Proton Co
mputed Tomography\nLocation: https://goo.gl/maps/XYQ7RgmwhowMYpUcA\nAbstra
ct: The high cost and low image quality traditionally associated with prot
on computed tomography (pCT) have prevented it from seeing significant use
in clinical settings. A cheap, compact, high-density scintillating glass
detector capable of being attached to existing proton therapy gantries may
help address these concerns. The design of the detector allows for use in
conjunction with single-proton counting reconstruction algorithms, as wel
l as beam-based algorithms that do not resolve individual protons within a
n accelerator bunch. This study presents quantitative reconstructed images
of proton stopping power from Monte Carlo generated pCT scans using the r
adiation transport code MCNP6, demonstrating the feasibility of proton ima
ging using this detector design. Relative error and contrast have been exa
mined and compared for images reconstructed using two reconstruction algor
ithms: a standard filtered back projection algorithm to act as a benchmark
, and a variant of a pCT algorithm which utilizes the concept of distance-
driven binning.
DTSTAMP:20220902T203621Z
END:VEVENT
BEGIN:VEVENT
SUMMARY:Symposium seminar: Review of modern smoothing methods for density a
nd distribution functions dealing with circular data from Dr. Yogen Chaube
y
DTSTART:20221124T213000Z
DTEND:20221124T223000Z
UID:56
LOCATION:https://goo.gl/maps/jdt4CQE4Nb62XXyN7
DESCRIPTION:Dr. Yogen Chaubey\nTitle: Review of modern sm
oothing methods for density and distribution functions dealing with circul
ar data\nLocation: https://goo.gl/maps/jdt4CQE4Nb62XXyN7\nAbstract: In thi
s talk, I will provide a short review of modern smoothing methods for dens
ity and distribution functions dealing with circular data. The usefulness
of circular kernels for smooth density estimation in this context is highl
ighted and contrasted with smooth density estimation based on orthogonal s
eries. It is seen that the wrapped Cauchy kernel as a choice of circular k
ernel appears as a natural candidate as it has a close connection to ortho
gonal series density estimation on a unit circle. In the literature, the u
se of the von Mises circular kernel is investigated, which requires the nu
merical computation of the Bessel function. On the other hand, the wrapped
Cauchy kernel is much simpler to use. This adds further weight to the con
siderable role of the wrapped Cauchy distribution in circular statistics.
We also investigate some transformation-based methods in adapting the line
ar kernel density estimator to the circular data. This is very useful in p
ractice as widely available software on linear kernel density estimation c
an easily be adapted to the circular case. These simpler methods are illus
trated using some real data.
DTSTAMP:20221020T170121Z
END:VEVENT
BEGIN:VEVENT
SUMMARY:Symposium seminar: Model-Based Clustering of Three-Way Data from Dr
. Paul McNicholas
DTSTART:20230328T220000Z
DTEND:20230328T233000Z
UID:73
LOCATION:EME 1151
DESCRIPTION:Dr. Paul McNicholas\nTitle: Model-Based Clusteri
ng of Three-Way Data\nLocation: EME 1151\nAbstract: Some model-based clust
ering approaches for three-way data are discussed. These include approache
s for multivariate longitudinal data\, approaches that do not assume norma
lity\, and an approach for three-way data that are high dimensional. Sever
al examples are discussed\, including a motivating example on data from a
large Canadian autism study. Finally\, some future directions are discusse
d.
DTSTAMP:20230309T215328Z
END:VEVENT
END:VCALENDAR