- Random - Mailing list — Dipartimento di Matematica, UniPI

Fwd: Algebraic methods in probability
by Elia Bisi 18 Mar '26

18 Mar '26

Dear colleagues, The programme below will be of interest to people working in Markov chain mixing, KPZ, random matrices, integrable systems, etc. Best regards, Elia ————————— Elia Bisi Assistant Professor (RTT) University of Florence Department of Mathematics & Computer Science Personal homepage Florence Probability group ---------- Forwarded message ---------- > Dear all, > > With Alexei Borodin and Evita Nestoridi, we are organizing a three week summer program at the Simons Center for Geometry and Physics in Stony Brook NY on "Algebraic methods in probability". The program runs from July 3 - 24, 2026 (this summer) and will have a mix of mini-courses (see the website https://scgp.stonybrook.edu/archives/45985) research lectures and lots of time for discussion and collaboration. This is a relaxed setting and the Simon center is able to provide housing and have a limited budget for travel (when people do not have their own sources). We expect to have about 80 people in total, with a mix of senior and junior faculty, as well as postdocs and graduate students. > > Applications are open until April 1. If you are interested, please apply now. Participation slots will quickly be filled after the April 1 deadline. > > Best, > Ivan Corwin

1 0

New Seminar Series: Rome Joint Seminars on Stochastic Dynamics Control and Learning
by Marta Leocata 17 Mar '26

17 Mar '26

Dear Colleagues, We are excited to announce a new cycle of Rome Joint Seminars on Stochastic Dynamics, Control, and Learning (with the acronym RoJo-SDCL), dedicated to the latest developments in stochastic modeling at the intersection of control theory, mean field games, reinforcement learning, and mathematical statistics, with applications in economics, finance, and climate science. The goal is to create a regular meeting point for applied mathematicians (and beyond), strengthening the research network in Rome and encouraging exchanges across universities and disciplines. Alongside the scientific presentations, we hope to have informal discussions to spark new collaborations and joint projects. The seminars will be held monthly, with two speakers per meeting. They will take place from 16:00 to 18:00 in the Conference Room of the EIEF–World Bank shared building (EIEF – Einaudi Institute for Economics and Finance), Via degli Abruzzi 10, Rome The schedule for this semester is as follows: - March 26, 2026: Yufei Zhang & Rüdiger Frey - April 16, 2026: Filippo Santambrogio & Daria Ghilli - May 14, 2026: Domenico Marinucci & Anna Vidotto Further information can be found at the following link: https://sites.google.com/view/rome-sdcl/home We look forward to seeing you there, and feel free to share this invitation with anyone who might be interested! Best regards, Valeria Bignozzi, Alessandro Bondi, Katia Colaneri, Jodi Dianetti, Marta Leocata

1 0

OWABI - Larry Wasserman - 25th March 2pm UK time
by Umberto Picchini 17 Mar '26

17 Mar '26

Dear all, the next OneWorld Approximate Bayesian Inference seminar www.warwick.ac.uk/owabi <http://www.warwick.ac.uk/owabi> is quickly approaching! We are delighted to announce that our next speaker is Larry Wasserman <https://www.stat.cmu.edu/~larry/><https://www.stat.cmu.edu/~larry/>(Carnegie Mellon University) who will talk about "/Robust Simulation Based Inference/" on *Wednesday the 25^th March at 2pm Uk time* (note the different day and time!) with an abstract reported below. The talk will be streamed at the OWABI MS Team channel General | OWABI Seminar: One World Approximate Bayesian Inference Seminar | Microsoft Teams <https://teams.microsoft.com/l/team/19%3AdhZ_4e_XLNJzCXPAMzTvT6BZ5KShEETkd_w…> at the following link *Join: *_https://teams.microsoft.com/meet/3560358333600?p=Klx01mfhGgKETxxz6K <https://teams.microsoft.com/meet/3560358333600?p=Klx01mfhGgKETxxz6K>_ Meeting ID: 356 035 833 360 0 Passcode: Fe3m7ty9 *Speaker*: Larry Wasserman <https://www.stat.cmu.edu/~larry/> (Carnegie Mellon University) *Title*: Robust Simulation Based Inference *Abstract*: Simulation-Based Inference (SBI) is an approach to statistical inference where simulations from an assumed model are used to construct estimators and confidence sets. SBI is often used when the likelihood is intractable and to construct confidence sets that do not rely on asymptotic methods or regularity conditions. Traditional SBI methods assume that the model is correct, but, as always, this can lead to invalid inference when the model is misspecified. This paper introduces robust methods that allow for valid frequentist inference in the presence of model misspecification. We propose a framework where the target of inference is a projection parameter that minimizes a discrepancy between the true distribution and the assumed model. The method guarantees valid inference, even when the model is incorrectly specified and even if the standard regularity conditions fail. Alternatively, we introduce model expansion through exponential tilting as another way to account for model misspecification. We also develop an SBI based goodness-of-fit test to detect model misspecification. Finally, we propose two ideas that are useful in the SBI framework beyond robust inference: an SBI based method to obtain closed form approximations of intractable models and an active learning approach to more efficiently sample the parameter space. *Keywords*: Exponential tilting, model misspecification, robust inference, simulation based inference, valid inference. *Reference*: L. Tomaselli, V. Ventura, L. Wasserman. Robust Simulation Based Inference. Preprint at ArXiv:2508.02404, <https://arxiv.org/pdf/2508.02404>2025. Best, Umberto Picchini on the behalf of the OWABI Organisers -- ____________________________________________________________________ Umberto Picchini, Full Professor of Mathematical Statistics https://umbertopicchini.github.io/ , Bluesky: @upicchini.bsky.social

1 0

SEMINAR: Alessandra Luati @Luiss - 26 mar
by Matteo Iacopini 17 Mar '26

17 Mar '26

Dear all, On Thursday, 26th March, at 12:00 in room A307b, the Department of AI, Data and Decision Sciences of Luiss University of Rome will host a seminar by Alessandra Luati (Imperial College London) Title: "Unobserved components, approximate filters, and dynamic adaptive mixture models” Abstract: State estimation in unobserved component models with parameter uncertainty is traditionally performed through approximate filters, where Gaussian distributions with given moments are employed to replace otherwise intractable conditional densities. This paper re-examines signal-plus-noise models where parameter uncertainty is induced by a latent variable that may assume a fixed number of states. First, it is shown that, for these models, the approximate filters commonly adopted in the literature can be obtained as linear combinations of minimum variance linear unbiased estimators. Second, it is observed that they coincide with filters implied by a novel class of dynamic adaptive mixture models, where the parameters of a mixture of distributions evolve over time following a recursion that is based on the score of the one-step-ahead predictive distribution. Focusing on a robust specification, where the mixture components are Student’s t distributions, we prove existence, stationarity, and ergodicity of the data generating process as well as invertibility of the filter, and consistency and asymptotic normality of the maximum likelihood estimator of the static parameters. An application to energy spot prices is discussed, where the novel specification is compared with, and shown to outperform, robust score-driven filters and the related class of mixture autoregressive models. 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

Course on "Federated Learning in Bioinformatics"
by Azzimonti Laura 17 Mar '26

17 Mar '26

Dear All, We are pleased to announce that the course on Federated Learning in Bioinformatics<https://www.sib.swiss/training/course/20260427_FEDBX> of the Swiss Institute of Bioinformatics (SIB) is open for applications. The course will be held in Lugano on 27 April 2026. This course introduces federated learning—a decentralised machine learning approach that enables multiple parties to collaboratively train a shared model using their own data sources without sharing the data itself, thereby enhancing privacy and security. In this course, you will learn the main concepts of federated learning and how to apply it to bioinformatics, which includes practicing on developing an algorithm for single‑cell data. This course is designed for life scientists and bioinformaticians from academia or industry who are curious about using machine learning while keeping sensitive data secure. Further detailed information, prerequisites and application form are available here<https://www.sib.swiss/training/course/20260427_FEDBX>. Should you have any questions, please don’t hesitate to contact us at training(a)sib.swiss<https://training@sib.swiss/>. With kind regards, Laura Azzimonti ––––––––––––––––––––––––––––––––––––––––––––– Scuola universitaria professionale della Svizzera italiana Istituto Dalle Molle di studi sull’intelligenza artificiale IDSIA USI-SUPSI Laura Azzimonti, PhD Senior Researcher and Lecturer Polo universitario Lugano Via la Santa 1 CH – 6962 Lugano-Viganello Tel. +41 (0)58 666 66 44 laura.azzimonti(a)idsia.ch www.idsia.ch

1 0

OWABI - Larry Wasserman - 25th March 2pm UK time
by Massimiliano Tamborrino 17 Mar '26

17 Mar '26

Dear all, the next OWABI seminar www.warwick.ac.uk/owabi<http://www.warwick.ac.uk/owabi> is quickly approaching! We are delighted to announce that our next speaker is Larry Wasserman<https://www.stat.cmu.edu/~larry/> <https://www.stat.cmu.edu/~larry/> (Carnegie Mellon University) who will talk about "Robust Simulation Based Inference" on Wednesday the 25th March at 2pm Uk time (note the different day and time!) with an abstract reported below. The talk will be streamed at the OWABI MS Team channel General | OWABI Seminar: One World Approximate Bayesian Inference Seminar | Microsoft Teams<https://teams.microsoft.com/l/team/19%3AdhZ_4e_XLNJzCXPAMzTvT6BZ5KShEETkd_w…> at the following link Join: https://teams.microsoft.com/meet/3560358333600?p=Klx01mfhGgKETxxz6K Meeting ID: 356 035 833 360 0 Passcode: Fe3m7ty9 Speaker: Larry Wasserman<https://www.stat.cmu.edu/~larry/> (Carnegie Mellon University) Title: Robust Simulation Based Inference Abstract: Simulation-Based Inference (SBI) is an approach to statistical inference where simulations from an assumed model are used to construct estimators and confidence sets. SBI is often used when the likelihood is intractable and to construct confidence sets that do not rely on asymptotic methods or regularity conditions. Traditional SBI methods assume that the model is correct, but, as always, this can lead to invalid inference when the model is misspecified. This paper introduces robust methods that allow for valid frequentist inference in the presence of model misspecification. We propose a framework where the target of inference is a projection parameter that minimizes a discrepancy between the true distribution and the assumed model. The method guarantees valid inference, even when the model is incorrectly specified and even if the standard regularity conditions fail. Alternatively, we introduce model expansion through exponential tilting as another way to account for model misspecification. We also develop an SBI based goodness-of-fit test to detect model misspecification. Finally, we propose two ideas that are useful in the SBI framework beyond robust inference: an SBI based method to obtain closed form approximations of intractable models and an active learning approach to more efficiently sample the parameter space. Keywords: Exponential tilting, model misspecification, robust inference, simulation based inference, valid inference. Reference: L. Tomaselli, V. Ventura, L. Wasserman. Robust Simulation Based Inference. Preprint at ArXiv:2508.02404,<https://arxiv.org/pdf/2508.02404>2025. Best, Massimiliano on the behalf of the OWABI Organisers ------ Dr. Massimiliano Tamborrino Reader (Associate Professor) and WIHEA Fellow Department of Statistics University of Warwick https://warwick.ac.uk/tamborrino

1 0

Call for applications 2026 (PhD at IMT Lucca)
by Irene Crimaldi 17 Mar '26

17 Mar '26

Dear all, The call for applications to the PhD program in *Economics, Analytics and Decision Sciences *(EADS) is out. We have at least *6 fully funded positions. * *More positions* could be added before the end of the call. *Please, share the call and prompt potentially interested students to apply. * Here you can find the *EADS* *website*: https://eads.imtlucca.it/ Here you can find the *information on the application procedure*: https://eads.imtlucca.it/how-to-apply_1 Here you can find the *call*: https://www.imtlucca.it/en/economics-analytics-and-decision-sciences-eads-a… *Deadline for applying: June 3 * My best regards, Irene --

1 0

SPASS seminar: Nicos Georgiou - Wed 25 Mar at 14:00
by Elia Bisi 17 Mar '26

17 Mar '26

Dear Colleagues, We are pleased to invite you to the following SPASS seminar (jointly organised by UniPi, SNS, UniFi and UniSi): Speaker: Nicos Georgiou (University of Sussex) Title: The totally asymmetric simple exclusion process (TASEP) on Galton-Watson trees Date and time: Wed 25th March at 14:00 Place: Room 107, Centro Didattico Morgagni, University of Florence Join online: meet.google.com/esm-ijgd-wai On behalf of the organisers, Elia Bisi ********************* Abstract: We define a totally asymmetric simple exclusion process on trees, where the particles enter the tree from a reservoir attached to the root. Particles can only jump following edges on the tree and only if they are jumping away from the root, to a non-occupied vertex. We are interested in two different aspects of this particle system. First we want to investigate the current of particles across a given generation at a specific time, when we start from an initially empty initial condition. Second, we would like some current estimates along a ray. Both answers depend heavily on our choice of jump rates for the particles and different regimes of behaviour will be presented during the talk. ————————— Elia Bisi Assistant Professor (RTT) University of Florence Department of Mathematics & Computer Science Personal homepage Florence Probability group

1 0

Postdoc position in Computational Modeling and Spatial Tumor Vasculature Analysis - Humanitas University
by Silvia MORLACCHI 16 Mar '26

16 Mar '26

Dear all, I am sharing the following postdoctoral position of behalf of Prof. Renne (salvatore.renne(a)hunimed.eu <mailto:salvatore.renne@hunimed.eu>) at Humanitas University. Best regards, Silvia Morlacchi Postdoctoral Researcher in Computational Modeling and Spatial Tumor Vasculature Analysis - Humanitas University The VISTE study is pioneering a unique interdisciplinary approach at the intersection of engineering, mathematics, and oncology. Our research integrates: Advanced mathematical modeling (Gaussian Processes, Bayesian inference, fractal geometry) to predict drug distribution in tumors High-resolution image analysis to reconstruct 3D vascular networks from histological samples Integration of spatial omics and computational pipelines to connect biological data with predictive quantitative models We are recruiting a highly motivated Postdoctoral Researcher with strong expertise in mathematics, computational modeling, and image analysis engineering to join this innovative effort in computational oncology. This position offers the opportunity to work at the frontier of cancer modeling, where engineering meets medicine and quantitative science drives clinical impact. For full details and application instructions, please see: https://jobs.humanitas.it/job/Pieve-Emanuele-Postdoctoral-Researcher-in-Com… and the project website https://spatialvesselanalysis.github.io/

1 0

Deep Patterns: from pattern formation to pattern recognition. Eurandom, May 18-22 2026
by Adriano Barra 16 Mar '26

16 Mar '26

Dear friends and colleagues, together with Wioletta Ruszel, Cristian Spitoni and Emilio N.M. Cirillo, we are pleased to announce the workshop "Deep Patterns: from pattern formation to pattern reconstruction": the event will take place at Eurandom, from May 18 2026 to May 22 2026. All the relevant information is available at the workshop website: https://www.eurandom.tue.nl/event/deep-patterns/ Thank you for your gentle attention to this matter. With our best regards, Adriano Barra (on the behalf of all the organizers)

1 0