BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Michigan Institute for Computational Discovery and Engineering - ECPv6.15.20//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Michigan Institute for Computational Discovery and Engineering
X-ORIGINAL-URL:https://micde.umich.edu
X-WR-CALDESC:Events for Michigan Institute for Computational Discovery and Engineering
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/Detroit
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20220313T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20221106T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20230312T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20231105T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20240310T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20241103T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20230921T110000
DTEND;TZID=America/Detroit:20230921T123000
DTSTAMP:20260605T143525
CREATED:20230914T172942Z
LAST-MODIFIED:20231018T162319Z
UID:10000643-1695294000-1695299400@micde.umich.edu
SUMMARY:SciML Webinar: David Ruhe - Geometric Clifford Algebra Networks
DESCRIPTION:Speaker: David Ruhe (University of Amsterdam) \n\n\nSession Chair: Erik Bekkers (University of Amsterdam) \n\nAbstract: In this talk\, I will present Clifford Group Equivariant Neural Networks\, an innovative method for building E(n)-equivariant networks based on Clifford (geometric) algebras. First\, I will give an introduction to the Clifford algebra and its geometric applications. Then\, I will introduce the Clifford group and how it always acts through the orthogonal group. As such\, a parameterization that is equivariant to the Clifford group will automatically be equivariant to the orthogonal group of\, e.g.\, rotations and reflections. We show that any polynomial (under the algebra’s geometric product) is such a parameterization.  We propose several layers from these insights and conduct experiments in three-\, four-\, and five-dimensional spaces. One of these experiments even includes equivariance to the nondefinite O(1\,3) Lorentz group from the same code implementation. Finally\, I will provide guidance on how to utilize our codebase for implementing these algorithms.
URL:https://micde.umich.edu/event/sciml-webinar-david-ruhe-geometric-clifford-algebra-networks/
CATEGORIES:SciML Webinar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20230928T110000
DTEND;TZID=America/Detroit:20230928T123000
DTSTAMP:20260605T143525
CREATED:20230918T023335Z
LAST-MODIFIED:20231018T163755Z
UID:10000648-1695898800-1695904200@micde.umich.edu
SUMMARY:SciML Webinar: Bob Carpenter - Multiscale Generalized Hamiltonian Monte Carlo with Delayed Rejection
DESCRIPTION:Speaker: Bob Carpenter (Flatiron Institute) \n\n\nSession Chair: Sam Livingstone (University College London) \n\n\nAbstract: In this talk\, I will demonstrate how we can combine two ideas\, generalized Hamiltonian Monte Carlo and delayed rejection\, to derive a sampler that is as efficient as Hamiltonian Monte Carlo\, but is able to adapt its step size to deal with multiscale distributions\, much like a standard integrator for ordinary differential equations. A distribution is multiscale if its curvature has different scales in the posterior; a textbook example is Radford Neal’s funnel example derived from hierarchical priors\, which has a very flat mouth (corresponding to high population variance) and very highly curved neck (low population variance). No fixed step size allows exploration of its posterior. Generalized HMC allows us to take a single Hamiltonian step along the gradient at a time (which is equivalent to Metropolis-adjusted Langevin dynamics)\, but only refresh momentum partially (which makes it underdamped). The naive form of this algorithm does not work because momentum must be reversed to maintain detailed balance if the Metropolis step rejects. To maintain directed exploration\, we apply delayed rejection\, which allows a proposal rejected due to divergence of the Hamiltonian (from too large a step size in the first-order approximation of the dynamics) to be retried with a smaller step size (with a Hastings-style adjustment for the retry). We show that the combination of delayed rejection and GHMC allows sampling multiscale distributions which otherwise lead to biased samples in standard Hamiltonian Monte Carlo (including dynamic forms such as the no-U-turn sampler). In conclusion\, I will discuss some preliminary work on applying the the automatic tuning method using complementary parallel chains developed by Matt Hoffman and Pavel Sountsov for their sampler MEADS (which also uses generalized HMC\, but with an alternative approach to maintaining directed exploration based on work of Radford neal\, which will also describe). \n\nSlides: https://statmodeling.stat.columbia.edu/wp-content/uploads/2023/09/carpenter-sciml-webinar-2023.pdf
URL:https://micde.umich.edu/event/sciml-webinar-bob-carpenter-multiscale-generalized-hamiltonian-monte-carlo-with-delayed-rejection/
CATEGORIES:Sciml,SciML Webinar Series
END:VEVENT
END:VCALENDAR