Dear colleagues,
it is our pleasure to invite you to the following seminar
Speaker: *Angelo Lucia*
Title: *Mixing time of quantum Gibbs samplers*
Time: Wednesday, *April 14*, 5PM (Italian time)
The abstract follows below in this message.
Zoom Link for the seminar
<https://us02web.zoom.us/j/86771523774?pwd=SitEVEhPSGRlaEkxTHZmQk9rTmlLZz09>
The seminar is a satellite activity of the PhD course "Introduction to
coercive inequalities with applications in analysis and probability theory"
(professor Boguslaw Zegarlinski - Imperial College, London)
Best regards,
Raffaella Carbone
Abstract.
Gibbs samplers are Markovian semigroups whose evolution converges
towards the termal equilibrium of a given Hamiltonian (the Gibbs
state). The time that the evolution takes, in the worst case, to
converge close to the thermal state is known as the mixing time.
Knowing the mixing time of a process is useful for many applications,
and it can be estimated with various functional inequalities.
In the case of classical spin systems on a lattice, it was shown that
it is possible to determine the mixing properties of the semigroup from
some "static" clustering condition of the thermal state.
This suggests the question of whether the same is true for quantum
spin systems.
In this talk, I will focus on the specific case of Gibbs states of
commuting Hamiltonians, and present some recent result in this
direction. I will introduce the notion of quantum conditional relative
entropy and show how it can be used to prove quasi-factorization (or
approximate tensorization) properties of the quantum relative entropy.
I will then show how these can be used to bound log-Sobolev constants
for product semigroups with heath-bath generators, and under stronger
assumptions in more general situations.
--
Raffaella Carbone, PhD
Probabilità e Statistica Matematica
Dipartimento di Matematica dell'Università degli Studi di Pavia
12-month postdoc fellowship from September 2021 (indicatively) at Université Laval (Quebec City), under the supervision of Marzia A. Cremona (Operation and Decision Systems Department) and Federico Severino (Finance Department).
Research area: Functional motif discovery for financial time series.
Requirements:
* Background in statistics, econometrics, or data science.
* PhD in related areas obtained less than 5 years ago.
* Proficiency in R and programming skills.
Primary goal:
Generalize and adapt existing statistical learning methods for motif discovery in functional data, to detect and predict bubbles and recurrent patterns in financial time series. An important part of the work is computational.
No teaching activity required.
Yearly salary: 51 737 CAD (maximum possible salary according to the postdoc union collective agreement).
Additional funds available for conference attendance.
A small amount of teaching assistantship can provide extra salary.
The candidate commits to move to Quebec City. In case public health conditions do not allow the moving, remote work can be considered.
For information or to apply, contact marzia.cremona(a)fsa.ulaval.ca<mailto:marzia.cremona@fsa.ulaval.ca> or federico.severino(a)fsa.ulaval.ca<mailto:federico.severino@fsa.ulaval.ca>
Marzia A Cremona
Department of Operations and Decision Systems
Université Laval
T 418 656-2131, poste 412525
https://marziacremona.com<https://marziacremona.com/>
Pavillon Palasis-Prince, local 2449
2325, rue de la Terrasse
Québec (Québec) G1V 0A6
Dear colleagues,
I would like to invite you to the following online seminar organized by the Probability group of the University of Pisa. The two talks will be accessible under the link
Click here to join the meeting<https://teams.microsoft.com/l/meetup-join/19%3Aaf3d635091e049579e555a84219a…>
Best regards,
Giacomo
Tuesday, April 13, 16:00
Speaker: Laure Dumaz (École Normale supérieure)
Title: Localization of the continuous Anderson hamiltonian in 1-d and its transition towards delocalization.
Abstract: We consider the continuous Schrödinger operator - d^2/d^x^2 + B’(x) on the interval [0,L] where the potential B’ is a white noise. We study the entire spectrum of this operator in the large L limit. We prove the joint convergence of the eigenvalues and of the eigenvectors and describe the limiting shape of the eigenvectors for all energies. When the energy is much smaller than L, we find that we are in the localized phase and the eigenvalues are distributed as a Poisson point process. The transition towards delocalization holds for large eigenvalues of order L. In this regime, we show the convergence at the level of operators. The limiting operator in the delocalized phase is acting on R^2-valued functions and is of the form ``J \partial_t + 2*2 noise matrix'' (where J is the matrix ((0, -1)(1, 0))), a form appearing as a conjecture by Edelman Sutton (2006) for limiting random matrices. Joint works with Cyril Labbé.
Tuesday, April 13, 17:00
Speaker: Martin Vogel (Université de Strasbourg)
Title: Eigenvalue asymptotics and eigenvector localization for non-Hermitian noisy Toeplitz matrices
Abstract: A most notable characteristic of non-Hermitian matrices is that their spectra can be intrinsically sensitive to tiny perturbation. Although this spectral instability causes the numerical analysis of their spectra to be extremely unreliable, it has recently been shown to be also the source of new mathematical phenomena. I will present recent results about the eigenvalues asymptotics and eigenvector localization for deterministic non-Hermitian Toeplitz matrices with small additive random perturbations. These results are related to recent developments in the theory of partial differential equations. The talk is based on joint work with J. Sjöstrand, and with A. Basak and O. Zeitouni.
************************
Giacomo Di Gesù
Dipartimento di Matematica
Università di Pisa
Largo Bruno Pontecorvo 5
56127 - Pisa, Italy
giacomo.digesu(a)unipi.it<mailto:giacomo.digesu@unipi.it>
https://sites.google.com/site/giacomodigesu/
Cari Colleghi
lunedì prossimo 12 Aprile dalle 16 alle 18 avrà luogo il prossimo
incontro per il ciclo di seminari Prisma. Gli speaker saranno Marco
Fuhrman e Andrea Cosso, qui sotto titoli, abstract ed il link al
collegamento Teams.
----------------------------------------------------------------
Speaker: Marco Fuhrman (Università di Milano)
Title:
A new tool in stochastic optimal control: the randomization method.
Abstract:
In the first part of this talk I will present a survery on the
relationships
among classical stochastic optimal control problems, non-linear partial
differential equations (the Hamilton-Jacobi-Bellman equations)
and backward stochastic differential equations (BSDEs).
In the second part, more specifically, I will introduce the so-called
randomization
method, which allows to associate an appropriate BSDE to a large class
of
optimal control problems. Among the possible generalizations,
I will concentrate on optimal control of path-dependent
equations, i.e. equations with general with memory effects.
Speaker: Andrea Cosso (Università di Bologna)
Title:
Randomization method and path-dependent Hamilton-Jacobi-Bellman equation
Abstract:
In the present talk I will study a stochastic optimal control problem
with path-dependent coefficients.
I will exploit the so-called randomization method to derive a dynamic
programming principle for the value function. This allows
to prove that the value function is a viscosity solution to a
path-dependent Hamilton-Jacobi-Bellman equation,
involving the horizontal and vertical derivatives of functional Ito
calculus.
Finally, I will discuss the validity of the comparison principle for
such a partial differential equation.
Collegamento Teams:
https://teams.microsoft.com/l/meetup-join/19%3a667d2414be564c5d8fba30acffeb…
---------------------------------------------------------------------
Grazie per l'attenzione, Domenico Marinucci e Claudia Ceci
WEBINARS IN STATISTICS @ COLLEGIO CARLO ALBERTO
<https://www.carloalberto.org/events/category/seminars/seminars-in-statistic…>
Venerdi 9 Aprile 2021, alle ore 12.00, si terrà il seguente webinar:
------------------------------------------------
Speaker: *Gilles Stupfler *(ENSAI Rennes and CREST, France)
Title: *Asymmetric least squares techniques for extreme risk estimation*
Zoom link:
https://us02web.zoom.us/j/81720792366?pwd=QWtaYVRRejVSVGlOMXFMaDdsemxjdz09
Meeting ID: 817 2079 2366
Passcode: 241135
Abstract:
Financial and actuarial risk assessment is typically based on the
computation of a single quantile (or Value-at-Risk). One drawback of
quantiles is that they only take into account the frequency of an extreme
event, and in particular do not give an idea of what the typical magnitude
of such an event would be. Another issue is that they do not induce a
coherent risk measure, which is a serious concern in actuarial and
financial applications. In this talk, I will explain how, starting from the
formulation of a quantile as the solution of an optimisation problem, one
may come up with two alternative families of risk measures, called
expectiles and extremiles. I will give a broad overview of their
properties, as well as of their estimation at extreme levels in
heavy-tailed models, and explain why they constitute sensible alternatives
for risk assessment using some real data applications. This is based on
joint work with Abdelaati Daouia, Irène Gijbels, Stéphane Girard and
Antoine Usseglio-Carleve.
------------------------------------------------
Il webinar è organizzato dalla "de Castro" Statistics Initiative
www.carloalberto.org/stats
Cordiali saluti,
Pierpaolo De Blasi
---
University of Torino & Collegio Carlo Alberto
carloalberto.org/pdeblasi
<https://sites.google.com/a/carloalberto.org/pdeblasi/>
-----------------------------------------------------------------------------
Concorso per un posto RTD-B a Milano Bicocca - MAT/06
-----------------------------------------------------------------------------
È stato bandito un posto RTD-B nel settore MAT/06 (Probabilità e Statistica
Matematica) presso il Dipartimento di Matematica e Applicazioni
dell'Università di Milano-Bicocca
Il bando si trova alla pagina
https://www.unimib.it/ateneo/gare-e-concorsi/2021-rtdb-042
La scadenza per la presentazione delle domande è il 29 aprile 2021.
Si prega di dare la massima diffusione presso tutti gli interessati.
--
Gianmario Tessitore
Dipartimento di Matematica e Applicazioni
Università degli Studi di Milano-Bicocca
*CALL FOR CONTRIBUTIONS***
On behalf of the Scientific and Organising Committees we are pleased to
inform you that contributions are now accepted for the Twelfth Workshop on
“Bayesian Inference in Stochastic Processes (BISP12)”
which will be held ONLINE on 27-28 May 2021 (afternoons, Central
European Summer Time).
Website:https://bisp.imati.cnr.it <https://bisp.imati.cnr.it> __
Attendance is free of charge but registration is required.
As in the past, the workshop will provide the opportunity to review,
discuss and explore directions of development of Bayesian inference in
stochastic processes. The workshop has always offered many young
researchers the opportunity to present their works in invited sessions.
This year the entire invited programme is dedicated to them. Ten
presentations by young scientists have been scheduled, followed by an
in-depth discussion by senior scholars in the field.
*We are now soliciting contributions for a virtual poster session. *
We plan to post contributed videos on the workshop website and keep them
there for a while after the event, with links to the presenting authors’
websites/emails, so that they can be contacted by interested people.
Unfortunately, time constraints will not allow for the presentation of
all those works during the workshop but we will allocate short spots for
ten contributors to highlight the main findings of their research and
few questions & answers.
An abstract must be submitted *by 18 April* via the BISP12 website,
where the abstract template (Word and Latex) and the guidelines for
submission are provided.
After submitting the abstract, contributors will receive an acceptance
notification email by *30 April* and instructions to upload a
pre-recorded video and/or a slide presentation by *15 May*. All
presentations will be published on the BISP12 website.
The workshop is organised by CNR-IMATI, Institute of Applied Mathematics
and Information Technology at the National Research Council of Italy,
Milano, www.imati.cnr.it <http://www.imati.cnr.it>.
Elisa Varini and Fabrizio Ruggeri
Chairs, Organising and Scientific Committees
*CALL FOR CONTRIBUTIONS*
On behalf of the Scientific and Organising Committees we are pleased to
inform you that contributions are now accepted for the Twelfth Workshop on
“*Bayesian Inference in Stochastic Processes (BISP12)*”
which will be held ONLINE on 27-28 May 2021 (afternoons, Central
European Summer Time).
_Websit__e_:bisp.imati.cnr.it__
Attendance is free of charge but registration is required.
As in the past, the workshop will provide the opportunity to review,
discuss and explore directions of development of Bayesian inference in
stochastic processes. The workshop has always offered many young
researchers the opportunity to present their works in invited sessions.
This year the entire invited programme is dedicated to them. Ten
presentations by young scientists have been scheduled, followed by an
in-depth discussion by senior scholars in the field.
*We are now soliciting contributions for a virtual poster session. *
We plan to post contributed videos on the workshop website and keep them
there for a while after the event, with links to the presenting authors’
websites/emails, so that they can be contacted by interested people.
Unfortunately, time constraints will not allow for the presentation of
all those works during the workshop but we will allocate short spots for
ten contributors to highlight the main findings of their research and
few questions & answers.
An abstract must be submitted *by 18 April* via the BISP12 website,
where the abstract template (Word and Latex) and the guidelines for
submission are provided.
After submitting the abstract, contributors will receive an acceptance
notification email by *30 April* and instructions to upload a
pre-recorded video and/or a slide presentation by *15 May*. All
presentations will be published on the BISP12 website.
The workshop is organised by CNR-IMATI, Institute of Applied Mathematics
and Information Technology at the National Research Council of Italy,
Milano, www.imati.cnr.it <http://www.imati.cnr.it>.
Elisa Varini and Fabrizio Ruggeri
Chairs, Organising and Scientific Committees