BEGIN:VCALENDAR
VERSION:2.0
PRODID:researchseminars.org
CALSCALE:GREGORIAN
X-WR-CALNAME:researchseminars.org
BEGIN:VEVENT
SUMMARY:Caroline Uhler (ETH/MIT)
DTSTART:20200605T130000Z
DTEND:20200605T140000Z
DTSTAMP:20260422T225657Z
UID:AlgebraicStatistics/1
DESCRIPTION:Title: Permutations and Posets for Causal Structure Discovery\nby
Caroline Uhler (ETH/MIT) as part of Algebraic Statistics Online Seminar\n
\nAbstract: TBA\n
LOCATION:https://researchseminars.org/talk/AlgebraicStatistics/1/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Bernd Sturmfels (MPI Leipzig/UC Berkeley)
DTSTART:20200619T190000Z
DTEND:20200619T200000Z
DTSTAMP:20260422T225657Z
UID:AlgebraicStatistics/2
DESCRIPTION:Title: Statistical Models with Rational Maximum Likelihood Estimator<
/a>\nby Bernd Sturmfels (MPI Leipzig/UC Berkeley) as part of Algebraic Sta
tistics Online Seminar\n\nAbstract: TBA\n
LOCATION:https://researchseminars.org/talk/AlgebraicStatistics/2/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Anna Seigal (University of Oxford)
DTSTART:20200703T130000Z
DTEND:20200703T140000Z
DTSTAMP:20260422T225657Z
UID:AlgebraicStatistics/3
DESCRIPTION:Title: Invariant theory for maximum likelihood estimation\nby Ann
a Seigal (University of Oxford) as part of Algebraic Statistics Online Sem
inar\n\nAbstract: TBA\n
LOCATION:https://researchseminars.org/talk/AlgebraicStatistics/3/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Ngoc Tran (UT Austin)
DTSTART:20200717T190000Z
DTEND:20200717T200000Z
DTSTAMP:20260422T225657Z
UID:AlgebraicStatistics/4
DESCRIPTION:Title: Graphical models for extreme events with tropical algebra\
nby Ngoc Tran (UT Austin) as part of Algebraic Statistics Online Seminar\n
\nAbstract: TBA\n
LOCATION:https://researchseminars.org/talk/AlgebraicStatistics/4/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Seth Sullivant (North Carolina State)
DTSTART:20200828T130000Z
DTEND:20200828T140000Z
DTSTAMP:20260422T225657Z
UID:AlgebraicStatistics/5
DESCRIPTION:Title: Identifiability in phylogenetics using algebraic matroids\
nby Seth Sullivant (North Carolina State) as part of Algebraic Statistics
Online Seminar\n\nAbstract: TBA\n
LOCATION:https://researchseminars.org/talk/AlgebraicStatistics/5/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Joe Kileel (UT Austin)
DTSTART:20200911T190000Z
DTEND:20200911T200000Z
DTSTAMP:20260422T225657Z
UID:AlgebraicStatistics/6
DESCRIPTION:Title: Fast symmetric tensor decomposition\nby Joe Kileel (UT Aus
tin) as part of Algebraic Statistics Online Seminar\n\nAbstract: TBA\n
LOCATION:https://researchseminars.org/talk/AlgebraicStatistics/6/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Thomas Kahle (OvGU Magdeburg)
DTSTART:20200925T130000Z
DTEND:20200925T140000Z
DTSTAMP:20260422T225657Z
UID:AlgebraicStatistics/7
DESCRIPTION:Title: Central limit theorems for permutation statistics\nby Thom
as Kahle (OvGU Magdeburg) as part of Algebraic Statistics Online Seminar\n
\nAbstract: TBA\n
LOCATION:https://researchseminars.org/talk/AlgebraicStatistics/7/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Serkan Hosten (San Francisco State University)
DTSTART:20201009T190000Z
DTEND:20201009T200000Z
DTSTAMP:20260422T225657Z
UID:AlgebraicStatistics/8
DESCRIPTION:Title: Two themes on (Gram) spectrahedra: central curves and symmetry
\nby Serkan Hosten (San Francisco State University) as part of Algebra
ic Statistics Online Seminar\n\n\nAbstract\nThis talk is based on two coll
aborations: one with Alex Heaton and Isabelle Shankar on symmetry adapted
Gram spectrahedra\, and the other with Isabelle Shankar and Angelica Torre
s on the degree of the central curve in semidefinite programming (SDP). Th
e objects in common are spectrahedra. The question for the degree of the
central curve in SDP (where feasible regions are spectrahedra) has its ans
wer in algebraic statistics as the ML degree of related linear concentrati
on models and the relevant geometry of complete quadrics. On the symmetry
adapted Gram spectrahedra side\, we use reductions in complexity of Gram s
pectrahedra for symmetric polynomials to understand the geometry of these
convex sets. Here I will focus on concrete families and examples.\n\nZoom
link: \nhttps://tum-conf.zoom.us/j/97632429442?pwd=RVNTb3NGb2t4QUdkSGQzelk
5S3luZz09\n
LOCATION:https://researchseminars.org/talk/AlgebraicStatistics/8/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Rina Foygel Barber (University of Chicago)
DTSTART:20201023T180000Z
DTEND:20201023T190000Z
DTSTAMP:20260422T225657Z
UID:AlgebraicStatistics/9
DESCRIPTION:Title: Testing goodness-of-fit and conditional independence with appr
oximate co-sufficient sampling\nby Rina Foygel Barber (University of C
hicago) as part of Algebraic Statistics Online Seminar\n\n\nAbstract\nGood
ness-of-fit (GoF) testing is ubiquitous in statistics\, with direct ties t
o model selection\, confidence interval construction\, conditional indepen
dence testing\, and multiple testing\, just to name a few applications. Wh
ile testing the GoF of a simple (point) null hypothesis provides an analys
t great flexibility in the choice of test statistic while still ensuring v
alidity\, most GoF tests for composite null hypotheses are far more constr
ained\, as the test statistic must have a tractable distribution over the
entire null model space. A notable exception is co-sufficient sampling (CS
S): resampling the data conditional on a sufficient statistic for the null
model guarantees valid GoF testing using any test statistic the analyst c
hooses. But CSS testing requires the null model to have a compact (in an i
nformation-theoretic sense) sufficient statistic\, which only holds for a
very limited class of models\; even for a null model as simple as logistic
regression\, CSS testing is powerless. In this paper\, we leverage the co
ncept of approximate sufficiency to generalize CSS testing to essentially
any parametric model with an asymptotically-efficient estimator\; we call
our extension “approximate CSS” (aCSS) testing. We quantify the finite
-sample Type I error inflation of aCSS testing and show that it is vanishi
ng under standard maximum likelihood asymptotics\, for any choice of test
statistic. We apply our proposed procedure both theoretically and in simul
ation to a number of models of interest to demonstrate its finite-sample T
ype I error and power. This work is joint with Lucas Janson.\n\nZoom link:
\nhttps://tum-conf.zoom.us/j/97632429442?pwd=RVNTb3NGb2t4QUdkSGQzelk5S3lu
Zz09\n
LOCATION:https://researchseminars.org/talk/AlgebraicStatistics/9/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Zehua Lai (University of Chicago)
DTSTART:20201106T200000Z
DTEND:20201106T210000Z
DTSTAMP:20260422T225657Z
UID:AlgebraicStatistics/10
DESCRIPTION:Title: Recht–Re Noncommutative Arithmetic-Geometric Mean Conjectur
e is False\nby Zehua Lai (University of Chicago) as part of Algebraic
Statistics Online Seminar\n\n\nAbstract\nStochastic optimization algorithm
s have become indispensable in modern machine learning. An important quest
ion in this area is the difference between with-replacement sampling and w
ithout-replacement sampling --- does the latter have superior convergence
rate compared to the former? A paper of Recht and Re reduces the problem t
o a noncommutative analogue of the arithmetic-geometric mean inequality wh
ere n positive numbers are replaced by n positive definite matrices. If th
is inequality holds for all n\, then without-replacement sampling (also kn
own as random reshuffling) indeed outperforms with-replacement sampling in
some important optimization problems. In this talk\, We will explain basi
c ideas and techniques in polynomial optimization and the theory of noncom
mutative Positivstellensatz\, which allows us to reduce the conjectured in
equality to a semidefinite program and the validity of the conjecture to c
ertain bounds for the optimum values. Finally\, we show that Recht--Re con
jecture is false as soon as n=5. We will also discuss some of the conseque
nces of our main theorem. This is a joint work with Lek-Heng Lim.\n\nZoom
link: https://tum-conf.zoom.us/j/97632429442?pwd=RVNTb3NGb2t4QUdkSGQzelk5S
3luZz09\n
LOCATION:https://researchseminars.org/talk/AlgebraicStatistics/10/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Kaie Kubjas (Aalto University)
DTSTART:20201120T140000Z
DTEND:20201120T150000Z
DTSTAMP:20260422T225657Z
UID:AlgebraicStatistics/11
DESCRIPTION:Title: Uniqueness of nonnegative matrix factorizations\nby Kaie
Kubjas (Aalto University) as part of Algebraic Statistics Online Seminar\n
\n\nAbstract\nNonnegative matrix factorizations are often encountered in d
ata mining applications where they are used to explain datasets by a small
number of parts. For many of these applications it is desirable that ther
e exists a unique nonnegative matrix factorization up to trivial modificat
ions given by scalings and permutations. This means that model parameters
are uniquely identifiable from the data. Different sufficient conditions f
or the uniqueness of nonnegative matrix factorizations have been well stud
ied\, however\, a little is known about necessary conditions. We will give
so far the strongest necessary condition for the uniqueness of a nonnegat
ive factorization. The talk is based on the joint work with Robert Krone.\
n\nZoom link:\n\nhttps://tum-conf.zoom.us/j/97632429442?pwd=RVNTb3NGb2t4QU
dkSGQzelk5S3luZz09\n
LOCATION:https://researchseminars.org/talk/AlgebraicStatistics/11/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Jose Rodriguez (University of Wisconsin - Madison)
DTSTART:20201204T200000Z
DTEND:20201204T210000Z
DTSTAMP:20260422T225657Z
UID:AlgebraicStatistics/12
DESCRIPTION:Title: Galois Groups in Statistics\nby Jose Rodriguez (Universit
y of Wisconsin - Madison) as part of Algebraic Statistics Online Seminar\n
\n\nAbstract\nSolving systems of polynomial equations is at the center of
applied algebraic geometry. A common theme of this field is to study a fam
ily of systems by allowing some of its coefficients to vary. In algebraic
statistics\, the role of coefficients is played by data and the solutions
we find yield maximum likelihood estimates\, critical points\, and/or impo
rtant information about a statistical model. An important invariant of a f
amily of systems is the Galois (monodromy) group. This captures important
symmetries within a system and has applications across kinematics\, comput
er vision\, power engineering and statistics. In each case\, the Galois gr
oup gives a description of how a system's solutions can vary with data.\n\
nIn this talk\, I will present three short stories about Galois groups app
earing in statistics. The first story emphasizes the idea of treating data
as coefficients of a polynomial system. We will visualize the monodromy
group acting in a nearest point problem where Euclidean distance (ED) degr
ees make an appearance. The next story involves Gaussian mixtures and deco
mposable systems. If time permits\, I will share a third story on how deco
mposable sparse systems play a role in solving the likelihood equations.\n
\nAttendees can join the seminar using the following Zoom link:\n\nhttps:/
/tum-conf.zoom.us/j/97632429442?pwd=RVNTb3NGb2t4QUdkSGQzelk5S3luZz09\n
LOCATION:https://researchseminars.org/talk/AlgebraicStatistics/12/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Eliana Duarte (OvGU Magdeburg)
DTSTART:20210118T130000Z
DTEND:20210118T140000Z
DTSTAMP:20260422T225657Z
UID:AlgebraicStatistics/13
DESCRIPTION:Title: Algebraic Geometry of Discrete Interventional Models\nby
Eliana Duarte (OvGU Magdeburg) as part of Algebraic Statistics Online Semi
nar\n\n\nAbstract\nThe Markov equivalence class of a discrete DAG model ca
n be described parametrically via the recursive factorization property or
implicitly by polynomial ideals which are defined via Markov properties of
the DAG. We address the problem of describing the Markov equivalence cla
sses of discrete DAG models with interventions using polynomial parameteri
zations and vanishing ideals. We show that the algebraic and combinatorial
properties of these models are captured via an interventional staged tree
model representation. This point of view leads us to a graphical characte
rization of the discrete interventional DAG models that are defined by bin
omial equations. This is joint work with Liam Solus (KTH\, Sweden)\, https
://arxiv.org/pdf/2012.03593.pdf.\n\nZoom link:\n\nhttps://tum-conf.zoom.us
/j/97632429442?pwd=RVNTb3NGb2t4QUdkSGQzelk5S3luZz09\n
LOCATION:https://researchseminars.org/talk/AlgebraicStatistics/13/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Wenxuan Guo (University of Chicago)
DTSTART:20210201T190000Z
DTEND:20210201T200000Z
DTSTAMP:20260422T225657Z
UID:AlgebraicStatistics/14
DESCRIPTION:Title: Shepp p-product\nby Wenxuan Guo (University of Chicago) a
s part of Algebraic Statistics Online Seminar\n\n\nAbstract\nIn 1962\, She
pp famously discovered a product of normal random variables that preserves
normality. The Shepp product\, which takes the form XY/(X^2 + Y^2)^1/2\,
has since been thoroughly studied and has found numerous connections to ot
her areas of statistics. Among other things\, it has an extension to n nor
mal variables\, gives a multiplicative analogue of central limit theorem\,
and applies unexpectedly to genomics as a test statistics for alignment-f
ree sequence analysis. The Shepp product is evidently the p = 2 special ca
se of XY/(X^p + Y^p)^1/p that we call the Shepp p-product. We will show th
at the Shepp p-product\, particularly when p = 1 and ∞ (the latter in a
limiting sense)\, is no less fascinating and applicable than the original
p = 2 case. Just as the Shepp 2-product preserves normal distributions\, t
he Shepp 1-product preserves Cauchy distributions while the Shepp ∞-prod
uct preserves exponential distributions. In fact\, the converse is also tr
ue in an appropriate sense\, allowing us to characterize the Cauchy\, norm
al\, and exponential distributions as the unique distributions preserved b
y the Shepp p-product for p = 1\, 2\, ∞ respectively. We will study the
multiplicative analogue of infinite divisibility with respect to the Shepp
p-product\, establish an asymptotic theory for the Shepp p-product of n i
.i.d. random variables\, and estimate the rates of convergence in Kolmogor
ov distance. Alongside our study of convergence rates\, we define the doma
in of normal attraction of extremal distributions and establish a new rate
of uniform convergence to Frechet distribution and reverse Weibull distri
bution. Some of our results are new even for the p = 2 case. We will also
discuss new applications of the Shepp p-product in statistics\, computatio
nal biology\, and statistical physics. This is joint work with Lek-Heng Li
m.\n\nJoin by Zoom:\n\nhttps://tum-conf.zoom.us/j/97632429442?pwd=RVNTb3NG
b2t4QUdkSGQzelk5S3luZz09\n
LOCATION:https://researchseminars.org/talk/AlgebraicStatistics/14/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Fatemeh Mohammadi (Ghent University)
DTSTART:20210215T130000Z
DTEND:20210215T140000Z
DTSTAMP:20260422T225657Z
UID:AlgebraicStatistics/15
DESCRIPTION:Title: Geometry of conditional independence models with hidden varia
bles\nby Fatemeh Mohammadi (Ghent University) as part of Algebraic Sta
tistics Online Seminar\n\n\nAbstract\nConditional independence (CI) is an
important tool in statistical modeling\, as\, for example\, it gives a sta
tistical interpretation to graphical models. In general\, given a list of
dependencies among random variables\, it is difficult to say which constra
ints are implied by them. Moreover\, it is important to know what constrai
nts on the random variables are caused by hidden variables. On the other h
and\, the CI statements are corresponding to some determinantal conditions
on the tensor of joint probabilities of the observed random variables. He
nce\, the geometric analogue of the inference question relates to determin
antal varieties and their irreducible decompositions. I will demonstrate h
ow the decompositions of CI varieties lead to interesting algebraic and co
mbinatorial questions about point configurations in matroid theory and inc
idence geometry. This\, in particular\, leads to effective computational a
pproaches for decomposing more general determinantal varieties.\n\nZoom li
nk: https://tum-conf.zoom.us/j/97632429442?pwd=RVNTb3NGb2t4QUdkSGQzelk5S3l
uZz09\n
LOCATION:https://researchseminars.org/talk/AlgebraicStatistics/15/
END:VEVENT
END:VCALENDAR