BEGIN:VCALENDAR
VERSION:2.0
PRODID:researchseminars.org
CALSCALE:GREGORIAN
X-WR-CALNAME:researchseminars.org
BEGIN:VEVENT
SUMMARY:Edgar Costa (MIT)
DTSTART:20211203T190000Z
DTEND:20211203T200000Z
DTSTAMP:20260422T212927Z
UID:WMmathcolloq/1
DESCRIPTION:Title: <a href="https://researchseminars.org/talk/WMmathcolloq
 /1/">The L-functions and modular forms database</a>\nby Edgar Costa (MIT) 
 as part of William & Mary Mathematics Colloquium\n\n\nAbstract\nThe Langla
 nds program\, first formulated by Robert Langlands in the 1960s and since 
 much developed and refined\, is a web of widespread conjectures that lie i
 n deep theories of mathematical symmetry\, it gives schematic direction to
  navigate between a dizzying array of subfields of mathematics\, including
  number theory\, representation theory\, algebraic geometry\, and harmonic
  analysis\nIn this talk\, we will explore some of these objects and connec
 tions and introduce the L-Functions and modular forms database (LMFDB) at 
 <a href="https://www.lmfdb.org/">https://www.lmfdb.org/</a>\, as one of it
 s goals is to provide a compelling visualization for the connections predi
 cted by the Langlands program.\nFor example\, we will see how the Riemann 
 hypothesis\, Goldbach's conjecture\, and Fermat's last theorem are related
 .\n
LOCATION:https://researchseminars.org/talk/WMmathcolloq/1/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Herbert Chang (University of Southern California)
DTSTART:20220225T190000Z
DTEND:20220225T200000Z
DTSTAMP:20260422T212927Z
UID:WMmathcolloq/2
DESCRIPTION:Title: <a href="https://researchseminars.org/talk/WMmathcolloq
 /2/">Fields Medals Are Concentrated in Mathematical ‘Families’</a>\nby
  Herbert Chang (University of Southern California) as part of William & Ma
 ry Mathematics Colloquium\n\n\nAbstract\nThe Fields Medal\, often referred
  as the Nobel Prize of mathematics\, is awarded to no more than four mathe
 maticians under the age of 40\, every 4 years. In recent years\, its confe
 rral has come under scrutiny of math historians\, for rewarding the existi
 ng elite rather than its original goal of elevating under-represented math
 ematicians. Prior studies of elitism focus on citational practices while c
 haracterization of the structural forces that prevent access remain unclea
 r. Here we show the flow of elite mathematicians between countries and lin
 go-ethnic identity\, using network analysis and natural language processin
 g on 240\,000 mathematicians and their advisor–advisee relationships. We
  present quantitative evidence of how the Fields Medal helped integrate Ja
 pan after WWII\, through analysis of the elite circle formed around Fields
  Medalists. We show increases in pluralism among major countries\, though 
 Arabic\, African\, and East Asian identities remain under-represented at t
 he elite level. Our results demonstrate concerted efforts by academic comm
 ittees\, such as prize giving\, can either reinforce the existing elite or
  reshape its definition. We anticipate our methodology of academic genealo
 gical analysis can serve as a useful diagnostic for equity and systemic bi
 as within academic fields. The presentation will also briefly discuss the 
 similar use of network science and graph theory in profiling misinformatio
 n during the Covid-19 pandemic.\n
LOCATION:https://researchseminars.org/talk/WMmathcolloq/2/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Dana P. Williams (Dartmouth College)
DTSTART:20220304T190000Z
DTEND:20220304T200000Z
DTSTAMP:20260422T212927Z
UID:WMmathcolloq/3
DESCRIPTION:Title: <a href="https://researchseminars.org/talk/WMmathcolloq
 /3/">The Equivalence Theorem for groupoid C*-algebras</a>\nby Dana P. Will
 iams (Dartmouth College) as part of William & Mary Mathematics Colloquium\
 n\n\nAbstract\nOne of the original motivations for the study of C*-algebra
 s came from noncommutative harmonic analysis and the group C*-algebra cons
 truction.  Nowadays the representation theory of C*-algebras is an interes
 ting subject onto itself.  An essential tool is the notion of Morita equiv
 alence of C*-algebras which is a good deal coarser than isomorphism\, but 
 still implies an equivalence of the representation theory.  There are many
  ways to build C*-algebras mimicing the group C*-algebra construction and 
 a key player is the construction of C*-algebras from groupoids.  Some time
  ago\, Jean Renault observed that a notion of groupoid equivalence implied
  Morita equivalence of the corresponding C*-algebras which gives a very co
 ncrete and topological way to establish deep analytic facts.  After briefl
 y outlining the necessary background\, I will give a sketch of this Equiva
 lence Theorem using a newer proof developed by Aidan Sims and myself.\n
LOCATION:https://researchseminars.org/talk/WMmathcolloq/3/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Mahlet Tadesse (Georgetown University)
DTSTART:20220325T180000Z
DTEND:20220325T190000Z
DTSTAMP:20260422T212927Z
UID:WMmathcolloq/4
DESCRIPTION:Title: <a href="https://researchseminars.org/talk/WMmathcolloq
 /4/">Uncovering cluster structures and relevant biomarkers in “-omics”
  data</a>\nby Mahlet Tadesse (Georgetown University) as part of William & 
 Mary Mathematics Colloquium\n\n\nAbstract\nHigh-throughput “-omics” te
 chnologies (genomics\, epigenomics\, transcriptomics\, proteomics\, metabo
 lomics\, etc) allow the simultaneous quantification of thousands of biomar
 kers. These technologies hold great potential for gaining insights into th
 e complex biological processes underlying specific phenotypes and for iden
 tifying biomarkers that can be used for improved diagnosis and therapeutic
  interventions. The challenges of analyzing the generated data have led to
  the development of various statistical\, computational and bioinformatic 
 tools over the the past couple of decades. In this talk\, I will present s
 ome of the methods we have proposed for uncovering cluster structures and 
 relevant biomarkers by combining ideas of mixture models and variable sele
 ction. I will discuss (1) a bi-clustering approach that allows clustering 
 on subsets of variables to refine disease classes and identify discriminat
 ing biomarkers\, (2) an integrative model to relate data from different -o
 mic levels using a stochastic partitioning method\, and (3) a mixture of r
 egression trees approach to uncover homogeneous disease subgroups and thei
 r associated predictors accounting for non-linear relationships and intera
 ction effects. I will illustrate the methods on various -omic studies.\n
LOCATION:https://researchseminars.org/talk/WMmathcolloq/4/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Tai Melcher (University of Virginia)
DTSTART:20220408T180000Z
DTEND:20220408T190000Z
DTSTAMP:20260422T212927Z
UID:WMmathcolloq/5
DESCRIPTION:Title: <a href="https://researchseminars.org/talk/WMmathcolloq
 /5/">Hypoellipticity in infinite dimensions</a>\nby Tai Melcher (Universit
 y of Virginia) as part of William & Mary Mathematics Colloquium\n\n\nAbstr
 act\nConsider a flat infinite surface to which one applies a unit of heat 
 at a fixed point and then steps away allowing the heat to propagate. The h
 eat flows quickly at first and then ever more slowly as it converges every
 where to zero. Contrast this now with the way heat flows on a curved surfa
 ce. For a short time\, the heat diffusion will be much the same\, but soon
  the geometry of the surface will play a role. For example\, if the surfac
 e has sharp corners\, this can create irregularities in the heat flow. If 
 the surface is smooth and closed\, like a ball\, the heat will eventually 
 travel back around to create complex interactions with itself\, converging
  to an equilibrium of uniformly positive heat everywhere.\n\nIt turns out 
 that the evolution of heat in a space is exactly related to the way a part
 icle randomly diffuses in that space subject to its geometry\, in the sens
 e that the probability that a random particle released from some initial p
 oint later finds itself inside some subset is exactly the proportion of he
 at in that subset from a unit of heat applied to the same initial point. A
 s suggested above\, nice properties of the geometry give rise to regularit
 y properties of these probabilities. More particularly\, in finite dimensi
 ons\, "hypoellipticity" is a standard assumption required for regularity. 
 Analogous regularity properties in infinite dimensions have allowed the de
 velopment of a calculus that has become an invaluable tool in the analysis
  of random processes and their applications. However\, in infinite dimensi
 ons it has remained elusive to demonstrate that hypoellipticity is a suffi
 cient condition for regularity. We will discuss some infinite-dimensional 
 model spaces where there have been positive results.\n\nIn addition to bei
 ng accessible on Zoom\, the talk will be projected in Jones 301.\n
LOCATION:https://researchseminars.org/talk/WMmathcolloq/5/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Ming-Jun Lai (University of Georgia)
DTSTART:20220311T190000Z
DTEND:20220311T200000Z
DTSTAMP:20260422T212927Z
UID:WMmathcolloq/6
DESCRIPTION:Title: <a href="https://researchseminars.org/talk/WMmathcolloq
 /6/">A Multivariate Spline based Collocation Method for Numerical Solution
  of PDEs</a>\nby Ming-Jun Lai (University of Georgia) as part of William &
  Mary Mathematics Colloquium\n\n\nAbstract\nThis talk is based on joint wo
 rk with Jinsil Lee. In this work\, we propose a collocation method based o
 n multivariate polynomial splines over triangulation/tetrahedralization fo
 r the numerical solution of partial differential equations. We start with 
 a detailed explanation of the method for the Poisson equation and then ext
 end the study to the second-order elliptic PDE in non-divergence form. We 
 shall show that the numerical solution can approximate the exact PDE solut
 ion very well under the assumption that the solution $u$ is in $H^3(\\Omeg
 a)$ over the domain $\\Omega$ which is of uniformly positive reach. Then w
 e present a large amount of numerical experimental results to demonstrate 
 the performance of the method over the 2D and 3D settings. In addition\, w
 e present a comparison with the existing multivariate spline methods to sh
 ow that the newly proposed method produces a similar and sometimes more ac
 curate approximation in a more efficient fashion.\n
LOCATION:https://researchseminars.org/talk/WMmathcolloq/6/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Jing Wang (University of Illinois Chicago)
DTSTART:20220422T180000Z
DTEND:20220422T190000Z
DTSTAMP:20260422T212927Z
UID:WMmathcolloq/7
DESCRIPTION:by Jing Wang (University of Illinois Chicago) as part of Willi
 am & Mary Mathematics Colloquium\n\nAbstract: TBA\n
LOCATION:https://researchseminars.org/talk/WMmathcolloq/7/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Yehua Li (UC Riverside)
DTSTART:20220429T180000Z
DTEND:20220429T190000Z
DTSTAMP:20260422T212927Z
UID:WMmathcolloq/8
DESCRIPTION:by Yehua Li (UC Riverside) as part of William & Mary Mathemati
 cs Colloquium\n\nAbstract: TBA\n
LOCATION:https://researchseminars.org/talk/WMmathcolloq/8/
END:VEVENT
END:VCALENDAR
