- Random - Mailing list — Dipartimento di Matematica, UniPI

PhD in probabilità a Warwick
by Giuseppe Cannizzaro 09 Oct '24

09 Oct '24

Cari colleghi, Il gruppo di probabilità di Warwick offre borse di studio per dottorati di ricerca. Qualora conosceste qualche bravo studente di master interessato, vi pregherei di far loro presente questa possibilità e inoltrare loro la mail. Cordiali saluti, Giuseppe Cannizzaro --------------------------------- The Probability group at the University of Warwick (*P@W <https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwarwick.a…>*) is looking for talented students to undertake their *PhD studies* with us. The successful applicant will join one of the *largest probability groups <https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwarwick.a…>* worldwide with an established track record of high-quality research and a broad range of interests <https://warwick.ac.uk/fac/sci/statistics/research/probability-at-warwick/re…>. The Probability group is *joint *between the Mathematics and Statistics Department and you can apply to either or *both* (recommended). See P@W <https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwarwick.a…> for a list of potential supervisors and their affiliation; *co-supervision* is possible and encouraged. The appointments include · *fully funded scholarships* for both UK and international students of the duration of 4 years; · a *travel allowance* for schools, workshops and conferences; · a *training fund* for extended research stays. Applications are received and assessed throughout the year on a *first-come-first-served basis*, with the prestigious Chancellor scholarship has a deadline of *9th December 2024. *For more information on funding please visit the webpage at the following *link <https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwarwick.a…>* . The programme consists of 6 months of *research training* with the possibility to attend graduate level courses taught by leading experts. Further, it is enriched by a variety of activities: 3 dedicated weekly *seminars*; *research focus groups* on cutting-edge topics; minicourses, workshops and conferences, ensuring a continued presence of international and distinguished visitors. Useful links: - P@W group: https://warwick.ac.uk/fac/sci/statistics/research/probability-at-warwick/ <https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwarwick.a…> - Research at P@W: https://warwick.ac.uk/fac/sci/statistics/research/probability-at-warwick/re… <https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwarwick.a…> - Fundings: https://warwick.ac.uk/fac/sci/statistics/postgrad/centre-for-doctoral-train… <https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwarwick.a…> - CDT in Statistics: https://warwick.ac.uk/fac/sci/statistics/postgrad/centre-for-doctoral-train… - CDT in Mathematics: https://warwick.ac.uk/fac/sci/maths/currentstudents/currentpostgraduates/ph…

1 0

SPASS seminar: Omiros Papaspiliopoulos
by Andrea Agazzi 09 Oct '24

09 Oct '24

Dear Colleagues, We would like to invite you to the following SPASS seminar, jointly organized by UniPi, SNS, UniFi and UniSi (abstract below): *Scalable large scale variational inference * by *Omiros Papaspiliopoulos* (Università Bocconi) The seminar will take place on *TUE, 15.10.2024* at *14:00 CET *in Aula Seminari, Dipartimento di Matematica, UNIPI and streamed online at this link <https://meet.google.com/gji-phwo-vbg>. The organizers, A. Agazzi, G. Bet, A. Caraceni, F. Grotto, G. Zanco https://sites.google.com/unipi.it/spass <https://www.google.com/url?q=https://sites.google.com/unipi.it/spass&source…> -------------------------------------------- *Title: Scalable large scale variational inference * *Abstract: The motivation behind this work is a computational framework for approximate statistical inference for the broad class of generalized bi-linear mixed models, special cases of which are generalized linear mixed models, factor models and probabilistic factorization models to mention a few popular examples. The objective is that the resultant inference has theoretically provable guarantees in terms of both uncertainty quantification and computational complexity. * * In this talk, I will first give some highlights of our work in using this framework for analyzing political ideology in Europe and then I will give a stand-alone short introduction to variational inference, discussing some theoretical aspects of its scalability.*

1 0

call for visiting professors
by Irene Crimaldi 09 Oct '24

09 Oct '24

Dear Colleagues, The call for visiting professors is out. Here are the links: https://www.imtlucca.it/it/jobopportunity/visiting-professor-positions-econ… https://inomics.com/job/visiting-professor-programme-in-economics-statistic… Irene Crimaldi

1 1

Seminario di Bertram Duering in Sapienza
by Enrico Scalas 08 Oct '24

08 Oct '24

Car* collegh*, Il 19 novembre 2024, in aula 24 del Dipartimento di Scienze Statistiche, Sapienza, Università di Roma, alle ore 14, il Prof. Bertram Düring della University of Warwick, Mathematics Institute terrà un seminario dal titolo From interacting agents to emerging patterns in economics and sociology Riassunto: Real-world systems in economics and sociology often consist of a large number of interacting agents that may form patterns or develop a collective behaviour. Typical applications include wealth and income distribution in an economy and opinion formation in a society. Mathematical techniques originating from statistical mechanics can be successfully developed further to model such systems and obtain mesoscopic and macroscopic descriptions of the microscopic dynamics. Such approaches frequently lead to generalisations of the classical Boltzmann equation for gas dynamics and associated limit partial differential equations. In this talk I will discuss different kinetic models for wealth distribution and opinion formation, and present analytical and numerical results. Tutte le persone interessate sono invitate a partecipare. Chi fosse interessato a partecipare a distanza, per favore, mi scriva a: enrico.scalas(a)uniroma1.it Cordiali saluti, Enrico Scalas

1 0

Uncertainty and Risk in Financial Mathematics, Milan Decembre 18-19
by Marco Maggis 08 Oct '24

08 Oct '24

Dear Colleagues, We are pleased to invite you to the workshop *"Uncertainty and Risk in Financial Mathematics,"* which will take place on *December 18-19, 2024* at the *Department of Mathematics, University of Milan, via C. Saldini 50*. This workshop will bring together academics and practitioners in the field to discuss the latest developments and challenges related to uncertainty and risk in financial mathematics. The confirmed speakers include: - Peter Bank - Anna Battauz - Fabio Bellini - Fulvia Confortola - Marzia De Donno - Alessandro Doldi - Sigrid Källblad - Athena Picarelli - Sergio Pulido - Nizar Touzi *Registration is free but mandatory* for organizational purposes. You can find more information and updates on the workshop website: Workshop Website <https://sites.google.com/view/unrifim24/home?authuser=0> We hope you will join us for this event, and we look forward to welcoming you to Milan. Best regards, Marco Maggis, Marco Frittelli, and Luciano Campi

1 0

post-doctoral positions in Paris - PSL
by CRISTINA TONINELLI 08 Oct '24

08 Oct '24

Dear colleagues, PSL university is awarding two Junior Research Chairs (three-year highend postdoctoral positions) as part of its research programme 'Statistical Physics and Mathematics’. This programme gathers physicists and mathematicians from three departments: * CEREMADE at University Paris Dauphine – PSL, * DMA at Ecole normale supérieure – PSL * LPENS at Ecole normale supérieure – PSL. Application deadline: 28th November 2024. Further details about the position and the application procedure can be found here : https://academicjobsonline.org/ajo/jobs/28737 Further details about our project and associated activities; may be found here https://www.ceremade.dauphine.fr/dokuwiki/psl-spm:start Feel free to also get in touch with one of us for any additional questions. WIth best wishes, the coordinators of the program Guillaume Barraquand Djalil Chafai Cristina Toninelli --------------------------------------------------------------------------------------------- Cristina Toninelli Postal Address: Ceremade Univ.Paris Dauphine Place du Marechal de Lattre de Tassigny 75775 Paris Cedex 16 - France Office: P207bis --------------------------------------------------------------------------------------------

1 0

Fwd: Post-doctoral position in Paris
by Maurizia Rossi 08 Oct '24

08 Oct '24

Dear all, Please find below more information about this position. Best wishes, Maurizia - - Maurizia Rossi Dipartimento di Matematica e Applicazioni Università degli Studi di Milano-Bicocca https://mauriziarossi.wordpress.com ---------- Messaggio inoltrato --------- Da: Nathael Gozlan <Nathael.Gozlan(a)math.cnrs.fr> Data: mar 8 ott 2024 alle 10:37 Oggetto: Post-doctoral position in Paris A: Natael Gozlan <Nathael.Gozlan(a)math.cnrs.fr> CC: Dear colleagues, A two years post-doctoral position is open in Université Paris Cité (Laboratoire MAP 5). It is funded by the ANR project SOCOT (Stochastic Orders and Constrained Optimal Transport). All candidates with a background in optimal transport, stochastic mass transport, convex geometry or functional inequalities are encouraged to apply. The position is expected to start in September 2025 or later. All relevant informations may be found in the pdf file attached to this email. The application deadline is January 31st. If you know young researchers who could be interested, please forward this email to them. Best regards, Nathael Gozlan

1 0

Seminario PRISMA oggi
by Francesco Caravenna 08 Oct '24

08 Oct '24

Seminari on-line del gruppo UMI - PRISMA (http://www.umi-prisma.polito.it/) Ricordiamo che oggi, martedì 8 ottobre 2024, Giovanni Conforti (Università di Padova) e Giacomo Greco (Università di Roma Tor Vergata) parleranno di Una passeggiata sui ponti di Schrödinger: dalla meccanica statistica all’algoritmo di Sinkhorn con il seguente orario: 16:00 Primo seminario 16:30 Pausa e discussione 16:45 Secondo seminario 17:15 Conclusione e discussione Trovate di seguito il riassunto. I seminari verranno trasmessi via Zoom al seguente link: https://uniroma1.zoom.us/j/88679569146 Meeting ID: 886 7956 9146 Vi aspettiamo numerosi! Valentina Cammarota e Francesco Caravenna %%%%%%%%%%%%%%%%%%%%%%%%%%%% RELATORI: Giovanni Conforti (Università di Padova) e Giacomo Greco (Università di Roma Tor Vergata) TITOLO: Una passeggiata sui ponti di Schrödinger: dalla meccanica statistica all’algoritmo di Sinkhorn. RIASSUNTO: Nel 1931, E.Schrödinger si interessò al problema di calcolare l’evoluzione più probabile di una nube di particelle Browniane indipendenti condizionatamente all’osservazione della loro configurazione agli istanti iniziale e finale. In linguaggio moderno, questo problema, detto problema di Schrödinger, viene formulato rigorosamente attraverso la teoria delle grandi deviazioni come un problema di minimizzazione entropica con vincoli marginali. Molti anni dopo, T.Mikami scoprì che il problema di Schrödinger converge verso il problema del trasporto ottimo, detto di Monge-Kantorovich, nel limite di rumore piccolo. Questo collegamento, che permette di vedere il problema di trasporto entropico come una regolarizzazione del problema deterministico, è stato largamente sfruttato negli ultimi anni nel contesto dello statistical machine learning. Il motivo principale di questo interesse è che il problema del trasporto entropico si può risolvere con un algoritmo veloce e di facile implementazione, l’algoritmo Sinkhorn. Tale algoritmo, introdotto negli anni '60, richiede solamente alcune semplici operazioni tra matrici, come la moltiplicazione componente per componente: in particolare, non è necessaria alcuna inversione di matrice. Dal punto di vista teorico, l’attrattiva principale risiede nella convergenza esponenziale nel numero di iterazioni. Tuttavia, la comprensione di questo fenomeno è ancora incompleta, per esempio nel caso base del costo quadratico, e più in generale per problemi di trasporto con costi non limitati e marginali a supporto non compatto. Anche per misure discrete, ci sono ancora numerose questioni aperte, per esempio riguardo alla dipendenza del tasso di convergenza dal parametro di regolarizzazione. L’obiettivo della prima parte di questo seminario è di introdurre il problema di Schrödinger come problema di meccanica statistica, esporre il collegamento con il problema di Monge Kantorovich, e introdurre l’algoritmo di Sinkhorn. Nella seconda parte il seminario si concentrerà sul problema della convergenza esponenziale, offrendo una panoramica su risultati più o meno recenti e le diverse tecniche di dimostrazione, e presentando alcune domande aperte. %%%%%%%%%%%%%%%%%%%%%%%%%%%% I seminari PRISMA hanno un formato di "colloquium" per creare un'occasione di scambio e discussione con tutta la comunità dei probabilisti e statistici italiani. Ogni giornata comprende due relatori che tengono due seminari di 30 minuti strettamente connessi, per presentare alla comunità una prospettiva sul proprio ambito di ricerca. Da quest'anno le registrazioni dei seminari vengono pubblicate sul canale YouTube dell'UMI: https://youtube.com/playlist?list=PLmySpc-jrtAMq84VH71evyqPc1hl6eEQb %%%%%%%%%%%%%%%%%%%%%%%%%%%%

1 0

Fwd: EPFL Statistics faculty search, and MSc
by Laura Maria Sangalli 07 Oct '24

07 Oct '24

Inoltro un avviso per posizioni e master presso EPFL. Cordialmente, Laura Sangalli —— Laura Maria Sangalli MOX - Dipartimento di Matematica Politecnico di Milano Piazza Leonardo da Vinci 32 20133 Milano - Italy (+39) 02 2399 4554 laura.sangalli(a)polimi.it<mailto:laura.sangalli@polimi.it> https://sangalli.faculty.polimi.it Inizio messaggio inoltrato: Da: Victor Panaretos <victor.panaretos(a)epfl.ch<mailto:victor.panaretos@epfl.ch>> Oggetto: EPFL Statistics faculty search, and MSc Data: 4 ottobre 2024, 16:16:56 CEST A: "Laura M. Sangalli" <laura.sangalli(a)polimi.it<mailto:laura.sangalli@polimi.it>> Rispondi a: <victor.panaretos(a)epfl.ch<mailto:victor.panaretos@epfl.ch>> Dear Laura, I'm writing to ask if you might be willing to disseminate some information related to statistics @ EPFL. First and foremost, we have an open rank faculty search (deadline November 1st) and we'd be grateful if you could spread the word among possibly interested colleagues: https://www.epfl.ch/about/working/page-26457-en-html/faculty-position-in-st… We are interested in any area of core statistics, whether it is related to methods, computation, theory, or applications! Second, EPFL has launched an MSc in Statistics open to students with a good mathematical background (https://www.epfl.ch/education/master/programs/statistics/) and we'd really appreciate it if you could communicate this to suitable students interested in further study abroad. Education at EPFL is very affordable, and it's in a beautiful region, so hopefully students will find it to be an attractive proposition. We will shortly send physical copies of a poster, and would be grateful if you could display them prominently in your department; a PDF version is here<https://www.dropbox.com/scl/fi/fcijzeep827ru7k33lo9o/master_of_statistics20…>, in case it would be useful. Of course we would be happy to return the favour both by mentioning your institution to our students and by displaying any posters you care to send. With my sincere thanks and best regards on behalf of the Statistics group at EPFL! Victor

1 0

mini-course on "Computability and Computational Complexity Questions in Dynamics"
by Stefano Marmi 07 Oct '24

07 Oct '24

Prof. Michael Yampolsky (University of Toronto) will give a series of 4 lectures on "Computability and Computational Complexity Questions in Dynamics" at the Scuola Normale Superiore in Pisa. The lectures will take place on Thursday October 24 and 31 and November 7 and 14, at 2:30 pm in Aula Volterra, in Piazza dei Cavalieri 7. The titles and abstracts of each of the 4 lectures are given below. Please contact me if you have any questions. Stefano Marmi stefano.marmi(a)sns.it 1) October 24: Computability and computational complexity in dynamics: can we trust numerical predictions of limiting behavior of orbits? The development of the modern subject of dynamical systems went hand-in-hand with numerical modeling. However, the theoretical basis of such modeling has remained largely unexplored and offers exciting challenges. What can and cannot be computed about the behavior of a dynamical system? I will give an overview of some of the recent results and directions of study. 2) October 31: Don’t believe your lying eyes: (non)-computability of Julia sets The talk will be a deep dive into the question of computability and computational complexity of Julia sets. Their fractal images are among the most-computed pictures in all of mathematics. But can we trust what these computations show? The results are surprising, and involve some beautiful complex analysis. The talk will be based on my work with M. Braverman, as well as more recent results obtained with A. Dudko, C. Rojas, and others. 3) November 7: How to lose at Monte Carlo: computing statistical behavior of dynamical systems Introduced by Ulam and von Neumann in the 1940s, Monte Carlo technique remains the most ubiquitous tool for statistical modeling. Surprisingly, our recent work with C. Rojas shows that such modeling provably fails for some very simple dynamical systems. I will present this theorem and discuss the resulting mathematical challenges. 4) November 14: Thurston versus Turing: algorithmic hardness of classifying finite dynamical systems Historically, the algorithmic aspects of classification problems have motivated much of the research in geometric topology (knot theory, classification of surfaces and 3-manifolds, etc). These questions are also very fruitful in dynamics. I will discuss the joint results with N. Selinger, K. Rafi and others on algorithmic decidability of Thurston equivalence and the connection with geometrization questions in dynamics.

1 0