May 2026 - Random - Mailing list — Dipartimento di Matematica, UniPI

SEMINAR: Petros Dellaportas @Luiss - 7 may
by Matteo Iacopini 05 May '26

05 May '26

Dear all, On Thursday, 7th May, at 12:00 in room A307b, the Department of AI, Data and Decision Sciences of Luiss University of Rome will host a seminar by Petros Dellaportas (University College London) Title: “Gaussian Invariant Markov Chain Monte Carlo” Abstract: We develop sampling methods, which consist of Gaussian invariant versions of random walk Metropolis (RWM), Metropolis adjusted Langevin algorithm (MALA) and second-order Hessian or Manifold MALA. Unlike standard RWM and MALA we show that Gaussian invariant sampling can lead to ergodic estimators with significantly improved statistical efficiency. This is due to a remarkable property of Gaussian invariance that allows us to obtain exact analytical solutions to the Poisson equation for Gaussian targets. These solutions can be used to construct efficient and easy-to-use control variates for variance reduction of estimators under any intractable target. We demonstrate the new samplers and estimators in several examples, including high-dimensional targets in latent Gaussian models, where we compare against several advanced methods and obtain state-of-the-art results. We also provide theoretical results regarding geometric ergodicity and an optimal scaling analysis that shows the dependence of the optimal acceptance rate on the Gaussianity of the target. Anyone interested in attending is welcome; just email me for logistical purposes. Hoping to meet you there! Best, Matteo Iacopini La presente e-mail proviene da Luiss Guido Carli e s'intende inviata per scopi lavorativi. Tutte le informazioni ivi contenute, compresi eventuali allegati, sono da ritenersi esclusivamente confidenziali e riservati secondo i termini del vigente D.Lgs. 196/2003 in materia di privacy e del Regolamento europeo 679/2016 - GDPR. È vietato qualsiasi ulteriore utilizzo non autorizzato. Qualora la stessa Le fosse pervenuta per errore, La preghiamo di eliminarla immediatamente e di darcene tempestiva comunicazione. Grazie. This e-mail message is sent by Luiss Guido Carli for business purposes. All informations contained therein, including any attachments, are for the sole use of the intended recipient and may contain confidential and privileged information pursuant to Legislative Decree 196/2003 and the European General Data Protection Regulation 679/2016 - GDPR -. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by soon reply this e-mail and destroy all copies of the original message. Thanks

1 0

Two seminars and a mini-course next week at Tor Vergata
by Michele Salvi 05 May '26

05 May '26

Dear all, Next week RoMaDS will host two seminars and a mini-course at the Department of Mathematics of the University of Rome, Tor Vergata (this is the link <https://www.mat.uniroma2.it/~rds/events.php> to RoMaDS webpage with all future events. We have some technical problems at the moment, but you can safely access the website ignoring the warning!). All abstracts can be found at the end of this email. Best, Michele > Monday, 11th May, Seminar Johannes Alt (University of Bonn) "Spectra of critical Erdős-Rényi graphs" 14h00-15h00, Department of Mathematics, Aula Dal Passo. Teams link <https://teams.microsoft.com/meet/35777624543024?p=boIC6CbJ8QfDutJzUV>. > Thursday, 14th May, Seminar Isabella Ziccardi (IRIF, Université Paris Cité) "Electing a Leader with Limited Resources" 16h00-17h00, Department of Mathematics, Aula D’Antoni. Teams link <https://teams.microsoft.com/meet/389989257555244?p=IMwo46pwx3w3fSzhOm>. > 12-13-14th May, Mini-course Cyril Letrouit (CNRS, Laboratoire de Mathématiques d'Orsay) "The Mathematics of Transformers: From Particle Systems to Control Theory" 12, 13, 14 May, 11h00-12h30, Department of Mathematics, Aula Dal Passo For all further information you can contact Prof. Caponigro at caponigro(a)mat.uniroma2.it <mailto:caponigro@mat.uniroma2.it> ABSTRACTS Abstract Alt “Spectra of critical Erdős-Rényi graphs" We consider the Erdős-Rényi graph G in its critical regime when its expected degree d scales like the logarithm of its number of vertices. On this critical scale, G undergoes a connectivity transition through the formation of isolated vertices. Moreover, localized eigenvectors emerge. The time evolution of a free quantum particle on G is governed by the adjacency matrix A of G through the Schrödinger equation. We determine the solution to this Schrödinger equation by comparison to an infinite tree. As A possesses localized and delocalized eigenvectors, the solution is in general a mixture of localized and scattering waves Abstract Ziccardi “Electing a Leader with Limited Resources” A distributed system consists of n independent entities—such as computers, servers, or even biological organisms—connected by a communication graph G. In such a system, nodes operate autonomously using their own local memory. A fundamental challenge in this setting is the Leader Election problem, where nodes must coordinate to elect a single leader. Electing a leader is fundamental because it serves as a central control, simplifying many other coordination tasks. Recently, much attention has been focused on leader election within weak communication models, like the beeping or stone age models. While many algorithms have been proposed, they typically require nodes to have large memory capacities, require knowledge of global quantities (such the total number of nodes n) or only work on specific families of graphs. In this talk, I introduce a very simple randomized algorithm that almost surely elects a leader on any network G. Our approach requires only constant memory and zero prior knowledge about G’s size or topology; nodes interact by sending only a 1-bit signal. We show that this protocol is efficient, reaching a stable leader configuration in O(D^2 log n) rounds with high probability, where D is the diameter of G. Finally, I will address the issue of resilience. Because our algorithm requires all nodes to start from the same, fixed initial state, it cannot guarantee leader election if the system starts in an arbitrary configuration reached due to local faults. I will discuss how to extend this algorithm to be self-stabilizing—meaning it can recover and converge from any initial state. Our self-stabilizing solution uses only O(log log n) bits of memory, representing an exponential improvement over previous literature that required O(log n) bits. This presentation is based on joint work with Robin Vacus [PODC 2025] and Lélia Blin and Sylvain Gay [PODC 2026] Abstract Letroiut “The Mathematics of Transformers: From Particle Systems to Control Theory” Since their introduction in 2017, Transformers have profoundly transformed large language models and, more generally, deep learning. This success largely relies on the mechanism known as "self-attention". In this course, I will introduce a mathematical framework that allows self-attention to be viewed as a system of interacting particles. I will explain certain remarkable properties of the associated dynamics in the space of probability measures, with particular emphasis on cluster formation, the preservation of Gaussian distributions, the subtleties of the associated mean-field limit, and the great "expressivity" of these neural networks, proved thanks to control theory.

1 0

PROMEMORIA MOX COLLOQUIUM presso Politecnico di Milano - 07.05.26 ore 14:30 – Peter Bühlmann
by Laura Maria Sangalli 05 May '26

05 May '26

Cari Colleghi, vi ricordo il seguente seminario della serie dei MOX COLLOQUIA: 07.05.26 Ore 14:30 - Sala Consiglio, Edificio 14, Politecnico di Milano Speaker: Peter Bühlmann, ETH Zürich Titolo: Causality-Inspired Machine Learning: A Path to more Robust AI Systems Abstract: The pursuit of reliable and robust machine learning has become a central focus, with statistics and data science playing pivotal roles in its advancement. We explore connections between distributional robustness and causality, providing methodological insights to enhance the reliability of AI systems. We examine the broader implications of these concepts through a case study in digital health and conclude by emphasizing the importance of rigorous real-world validation of machine learning and AI algorithms. https://mox.polimi.it/mox-colloquia-seminars-list/mox-seminars/?id_evento=2… <https://cassyni.com/events/Lwo25Mh9yCpmezYk7Fie8m> [NgwyDiUV254sqYXdcEuAM4.png] Causality-Inspired Machine Learning: A Path to more Robust AI Systems<https://cassyni.com/events/Lwo25Mh9yCpmezYk7Fie8m> cassyni.com<https://cassyni.com/events/Lwo25Mh9yCpmezYk7Fie8m> Per seguirlo in streaming è necessario registrarsi (gratuitamente) alla piattaforma Cassyni. Un caro saluto, 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 https://sangalli.faculty.polimi.it

1 0

EXTENDED DEADLINE for AMASES 2026
by Francesco Cesarone 04 May '26

04 May '26

Dear Colleagues, Due to numerous requests, the Organizing Committee of the 50th Annual Meeting of the Association for Mathematics Applied to Social and Economic Sciences (AMASES 2026) https://sites.google.com/uniroma1.it/amases2026/<https://urldefense.com/v3/__https:/sites.google.com/uniroma1.it/amases2026/…> has decided to extend the deadline for abstract submissions. New Deadline for Abstract Submission: May 10, 2026 Notification of Acceptance: June 5, 2026 Deadline for Early Bird Registration: June 30, 2026 Deadline Registration for Communications to be included in the program: June 30, 2026 Francesco Cesarone On behalf of The AMASES 2026 Organizing Committee * F Cesarone, M Corradini, L Lampariello (2024), Matematica generale, Edizioni Giappichelli, ISBN 9791221107746<https://www.giappichelli.it/matematica-generale-9791221107746> * F Cesarone, M Corradini, L Lampariello (2024), Esercitazioni di matematica generale, Edizioni Giappichelli, ISBN 9791221107753<https://www.giappichelli.it/esercitazioni-di-matematica-generale-9791221107…> * F Cesarone (2020), Computational Finance. MATLAB oriented modeling, Routledge-Giappichelli Studies in Business and Management, ISBN 978-0-367-49303-5<https://www.giappichelli.it/computational-finance-9788892132504> For more info: https://www.francescocesarone.com/books -- Francesco Cesarone - Ph.D. Full Professor Department of Business Studies Roma Tre University Via Silvio D'Amico, 77 00145 - Roma, Italy phone: +39 06 57335744 Skype: francesco.cesarone email: francesco.cesarone(a)uniroma3.it<mailto:francesco.cesarone@uniroma3.it> office no 16, 5th floor WWW: https://www.francescocesarone.com<https://www.francescocesarone.com/>/<https://www.francescocesarone.com/>

1 0

Reminder: 44th Finnish Summer School on Probability and Statistics, 50 years jubilee, Lammi, Finland 25-29.5.2026
by Gasbarra, Dario 04 May '26

04 May '26

Dear friends of Stochastics and Statistics, registration for the 44th Finnish Summer School on Probability and Statistics, 50 years jubilee edition, in Lammi, Finland 25-29.5.2026 is open until SATURDAY MAY 9th. Please check the website 44th Finnish Summer School on Probability and Statistics, 50 years jubilee | FDNSS<https://fdnss.fi/44th-finnish-summer-school-on-probability-and-statistics-5…> https://fdnss.fi/wp-content/uploads/2026/03/fdnss-school_poster-2026.pdf Minicourses: * Nils Berglund<https://www.idpoisson.fr/berglund/> (Université d’Orléans) Topics in Gaussian Wiener chaos expansion (abstract)<https://fdnss.fi/topics-in-gaussian-wiener-chaos-expansion/> * Giulia di Nunno<https://www.mn.uio.no/math/personer/vit/giulian/> (University of Oslo) BSDEs and fully-dynamic risk measures * Frank Röttger<https://sites.google.com/view/roettger/startseite> (University of Twente) Probabilistic graphical models and their application to extreme value statistics * SJS Lectures: Subhabrata Sen<https://sites.harvard.edu/sus977/> (Harvard University) Recent progress in multi-modal data analysis (abstract)<https://fdnss.fi/recent-progress-in-multi-modal-data-analysis/> See you in Lammi ! Yours Dario Gasbarra, Department of Mathematics and Statistics University of Vaasa, P.O.Box 700 65101 Vaasa Finland dario.gasbarra(a)uwasa.fi

1 0

Seminario Sapienza Roma - 7 Maggio 2026
by Alessandro Degregorio 04 May '26

04 May '26

Cari tutti, vorrei segnalare il seguente seminario che si terrà il *7 Maggio* *alle ore 14, aula 24,* presso il *Dipartimento di Scienze Statistiche -- Sapienza Roma * (Edificio CU002, IV piano, p.le A.Moro 5, Roma). *Speaker:* Kei Kobyashi (Fordham University - New York) *Title: *Small-time asymptotic behavior of the spectral heat content for isotropic Lévy processes *Abstract: *The spectral heat content (SHC) measures the total heat contained in a bounded domain at a given time when the initial temperature inside the domain is set to be one, and the temperature outside the domain is kept at zero. The SHC with heat particles represented by Brownian motion was intensively studied more than three decades ago, whereas the investigation of the SHC for Lévy processes started only about a decade ago. In this talk, we derive the precise small-time asymptotic behavior of the SHC for a large class of isotropic Lévy processes. The result covers the cases of Brownian motion, stable Lévy processes, and some jump diffusions, while the proof provides a unified approach to the asymptotic behavior of the SHC for all of these processes. This is joint work with Hyunchul Park. E' possibile assistere al seminario online collegandosi al seguente indirizzo: https://uniroma1.zoom.us/j/88659953902?pwd=bRglbpbc8ZyN4BBxllXPxlncAa1l6i.1 (ID riunione: 886 5995 3902) Cordiali saluti, Alessandro De Gregorio -- *Fai crescere le giovani ricercatrici e i giovani ricercatori*** *con il 5 per mille alla Sapienza* Scrivi il codice fiscale dell'Università *80209930587 **Cinque per mille <https://www.uniroma1.it/it/node/23149>*

1 0

SPASS seminar: Lucie Laurence - Tue 5 May at 14:00
by Marco Romito 03 May '26

03 May '26

Dear Colleagues, We are pleased to invite you to the following SPASS seminar <https://sites.google.com/unipi.it/spass/seminars?authuser=0> (jointly organised by UniPi, SNS, UniFi and UniSi): _Speaker_: *Lucie Laurence *(University of Bern, Institute of Mathematical Statistics and Actuarial Sciences) _Title_: *Noise-induced stabilization in chemical reaction networks* _Date and time_: *Tue, May 5th* at *14:00 CET* _Place_: Auletta Riunioni, Dipartimento di Matematica, UniPi, Pisa _Join online_: https://meet.google.com/esx-pxkw-fhw On behalf of the organisers, Samuele Saviozzi ********************* _Abstract_: /Chemical reaction networks (CRNs) describe the evolution of interacting molecular species through a set of reactions. In the stochastic formulation, the system tracks discrete molecule counts and evolves as a jump process driven by reaction events. Transition rates follow mass‑action kinetics and are polynomial functions of the state. When molecule numbers are large, these stochastic dynamics are typically approximated by a deterministic ODE system. The connection between the stability of these two models is an active area of research./ /In this talk, I will present a CRN exhibiting a noise‑induced stabilization phenomenon. Deterministically, the associated ODE admits trajectories that diverge and blow up in finite time. In contrast, the stochastic model is stable: sample paths remain bounded and repeatedly return to compact regions of the state space. The proof combines a Lyapunov‑function criterion with scaling arguments that control the behavior of the stochastic trajectories./ /In the final part of the talk, I will discuss new directions for extending this stabilization result to broader classes of CRNs. Using Poincaré compactification, I will outline conditions on the deterministic dynamics that ensure stability of the corresponding stochastic system./ /This is a joint work with A. Agazzi./

1 0