Seminari - Dipartimento Informatica validi dal 03.12.2025 al 03.12.2026. https://www.di.univr.it/?ent=seminario&rss=0 https://www.di.univr.it/?ent=seminario&rss=0&id=6843 Relatore: Elio La Rosa; Provenienza: MCMP at Ludwig-Maximilians-Universität München; Data inizio: 2025-12-03; Ora inizio: 15.00; Note orario: Sala Verde; Riassunto: Extensions of Intuitionistic logic with Hilbertrsquo;s epsilon terms are famously non-conservative over Intuitionistic quantification. Partial solutions rely on weaker axioms, extensions to existence predicate or stronger intermediate logics, but do not scale well to theories. In this talk, I present a new approach, implicit in early papers by Melvin Fitting on Herbrandrsquo;s theorem for modal logics. Fitting extends quantified S4 modal logic with epsilon terms and predicate abstraction, using lambda binders to create a scoping mechanism for terms that mimics that of quantifiers. The resulting logic is conservative over S4. An immediate but overlooked consequence is its faithful embedding of Intuitionistic quantifiers over their modal and epsilon translation. I show that this result can be improved by internalizing the S4 modalities directly into lambda abstractors, thus eliminating modalities from the language entirely. The resulting logic is still conservative extension over Intuitionistic quantification, and remains so when adding identity and term extensionality, opening the way for applications to constructive theories. Wed, 3 Dec 2025 15:00:00 +0100 https://www.di.univr.it/?ent=seminario&rss=0&id=6843 https://www.di.univr.it/?ent=seminario&rss=0&id=6824 Relatore: Alessio Merlo//Giacomo Longo; Provenienza:  University School of Advanced Defense Studies (Roma); Data inizio: 2025-12-09; Ora inizio: 14.00; Note orario: Sala Verde (presenza ed on line); Referente interno: Mariano Ceccato; Riassunto: Abstract :The digitalisation of transportation is turning ships and aircraft into complex systems driven by networks, software and sensors. This evolution brings new capabilities but also introduces weaknesses that can affect safety itself. The presentation explores recent research on cybersecurity across maritime and aviation domains, highlighting practical attacks, technological constraints and defence strategies designed for atypical and tightly regulated environments. Short Bios : Alessio Merlo (Ph.D. 2010) is a full professor in Computer Engineering and the Director of the CASD - University School of Advanced Defense Studies in Rome. He has been a researcher on cybersecurity since 2010, mainly focusing on mobile security, data privacy, and cyberphysical systems security. Giacomo Longo (Ph.D. 2025) is a postdoctoral researcher in Computer Engineering at CASD - University School of Advanced Defense Studies in Rome. His research interests revolve around cyber physical systems security, the applications of simulation technologies to cyber security, and virtualized sensing. Link Zoom: https://univr.zoom.us/j/85346717538. Tue, 9 Dec 2025 14:00:00 +0100 https://www.di.univr.it/?ent=seminario&rss=0&id=6824 https://www.di.univr.it/?ent=seminario&rss=0&id=6845 Relatore: Luca Pinello; Provenienza: Harvard University; Data inizio: 2025-12-09; Ora inizio: 14.30; Note orario: Aula T.06; Referente interno: Rosalba Giugno; Riassunto: Mini-course (6 hours) on quot;Advanced Algorithms to understand regulatory in personal genomicsquot; within the course quot;Multiomics Single cell Analysisquot; in the Master Programme in Medical Bioinformatics A.Y. 2025 - 2026 09/12/2025 - dalle 14:30 alle 17:30 - Aula T.06 10/12/2025 - dalle 8:30 alle 11:30 - Aula T.06. Tue, 9 Dec 2025 14:30:00 +0100 https://www.di.univr.it/?ent=seminario&rss=0&id=6845 https://www.di.univr.it/?ent=seminario&rss=0&id=6838 Relatore: Walter Boscheri; Provenienza: Université Savoie Mont Blanc and CNRS (France); Data inizio: 2025-12-10; Ora inizio: 9.30; Note orario: l'evento si ripete il 11/12/2025 alle 9:00; Referente interno: Elena Gaburro; Riassunto: This course focuses on numerical methods for solving multiscale partial differential equations, with applications to compressible fluid dynamics across different Mach number regimes. Implicit and explicit approaches are introduced through the advection-diffusion equation, followed by an analysis of the Euler equations for compressible gas dynamics in non-dimensional form and their asymptotic limits. The concept of Asymptotic Preserving (AP) schemes is explored through IMEX and semi-implicit formulations. A significant part of the course is devoted to a hands-on Matlab implementation of a semi-implicit scheme for the one-dimensional Euler equations. The course concludes with an outlook on advanced topics, including high-order IMEX methods and exactly divergence-free operators. Timetable: Wednesday 10/12/2025: room ALFA (9:30-12:30) Thursday 11/12/2025: room H (9:00 - 12:00) . Wed, 10 Dec 2025 09:30:00 +0100 https://www.di.univr.it/?ent=seminario&rss=0&id=6838 https://www.di.univr.it/?ent=seminario&rss=0&id=6802 Relatore: Riccardo Sartea; Provenienza: Amazon; Data inizio: 2025-12-10; Referente interno: Alberto Castellini; Riassunto: Mini-course (6 hours) on quot;Applied Large Language Models: Hands-onquot; within the course Natural Language Processing in the Master Programme in Artificial Intelligence A.Y. 2025 - 2026 Schedule: - 10/12: 8.30-10.30 Aula T02.a - 11/12: 12.30-14.30 Aula I (di Imola) - 12/12: 8.30-10.30 Aula T02.b Contact: Alberto Castellini (alberto.castellini@univr.it). Wed, 10 Dec 2025 00:00:00 +0100 https://www.di.univr.it/?ent=seminario&rss=0&id=6802 https://www.di.univr.it/?ent=seminario&rss=0&id=6846 Relatore: RIccardo Sartea; Provenienza: Amazon; Data inizio: 2025-12-12; Ora inizio: 14.00; Note orario: Sala Verde (presenza); Referente interno: Alberto Castellini; Riassunto: Abstract : How do we bridge the gap between raw data and automated business strategy? This session provides a technical deep dive into four distinct implementations of AI designed to solve this problem across the healthcare, insurance, and retail sectors. We will begin by detailing the specific NLP techniques required to identify medicines and pathologies within unstructured clinical narratives, followed by a look at how Generative AI is transforming information extraction in the insurance domain. The focus then shifts to the architecture of personalization systems, comparing Deep Learning approaches for product affinity against tree-based models for content prioritization. Finally, we will define the emerging role of the quot;Agentic Orchestrator,quot; a system that synthesizes these predictive insights to autonomously decide the optimal marketing engagement for every customer interaction. Bio : With a Ph.D. in Artificial Intelligence from the University of Verona, Riccardo Sartea bridges the gap between theoretical foundations and practical application. Currently a Data Scientist at Amazon Web Services Zurich, he focuses on architecting end-to-end machine learning solutions that scale from initial concept to production. His recent work centers heavily on the lifecycle of Large Language Models, applying techniques like fine-tuning and knowledge distillation to address diverse use cases. Beyond the algorithms, he collaborates closely with engineering and product teams to translate ambiguous challenges into concrete, data-driven technical strategies. Fri, 12 Dec 2025 14:00:00 +0100 https://www.di.univr.it/?ent=seminario&rss=0&id=6846 https://www.di.univr.it/?ent=seminario&rss=0&id=6726 Relatore: Dante Kalise; Provenienza: Imperial College London; Data inizio: 2025-12-15; Ora inizio: 14.30; Note orario: da definire; Referente interno: Giacomo Albi; Riassunto: This course introduces the fundamental ideas and computational methods behind optimal control and data-driven modelling.In this short course, we will study how to incorporate elements of machine learning into optimal control design. The course will focus on fundamentals on optimal control: dynamic optimization, linear-quadratic control, dynamic programming and Pontryagin#39;s maximum principle. Nonlinear optimal control. Approximation methods in high dimensions are also discussed such as polynomial approximation, deep neural networks. Optimization techniques: LASSO regression, stochastic gradient descent, training neural networks. Finally, combining the first two parts, we will study the construction of data-driven schemes for the approximation of high-dimensional nonlinear control laws. . Mon, 15 Dec 2025 14:30:00 +0100 https://www.di.univr.it/?ent=seminario&rss=0&id=6726 https://www.di.univr.it/?ent=seminario&rss=0&id=6734 Relatore: Marcello Carioni; Provenienza: University of Twente; Data inizio: 2025-12-16; Ora inizio: 8.30; Note orario: Aula G; Referente interno: Giacomo Albi; Riassunto: Optimal Transport (OT) is a mathematical theory introduced by Gaspard Monge in 1781 to study the optimal allocation of resources and goods. Its original formulation, known as the Monge formulation, aims at finding the best way to transport a probability distribution in $\mathcal{P}(\mathbb{R}^n)$ to another by minimizing a transport cost computed with respect to a given cost function $c$. Mathematically, this consists in finding a map $T : \mathbb{R}^n \to \mathbb{R}^n$ that minimizes the cost of moving mass from the first measure to the second one. This formulation is highly flexible, as it encompasses discrete, semi-discrete, and continuous settings, and the cost $c$ can be chosen to enforce specific properties, such as restricting transport to certain regions or promoting mass concentration. Due to its flexibility and mathematical rigor, the theory experienced substantial development in the 20th century and gained relevance in fields such as economics, urban planning, image processing, and biology. More recently---especially in the last decade---optimal transport has become a powerful tool in machine learning. Many modern algorithms rely on estimating distances between data distributions efficiently and accurately, and OT provides a natural framework for comparing such distributions by quantifying the cost of transporting one into the other. This perspective, combined with fast numerical methods for computing OT (notably entropic regularization and the Sinkhorn algorithm), has made OT central in the design of new generative models, in solving inverse problems, and in improving the robustness of neural networks. In this series of lectures, we will cover: (1) an introduction to the classical formulations of optimal transport and the core theoretical results; (2) entropic regularization and the Sinkhorn algorithm for efficient computation of regularized OT; and (3) connections between OT and machine learning, with a focus on adversarial generative models based on optimal transport (such as WGAN and WAE). Time permitting, we will also discuss OT-based approaches to inverse problems. Schedule: Tue 16/12 Aula G, 8:30-10:30 Wed 17/12 Aula Alfa, 13:30-15:30 Thu 18/12 Aula T.05, 10:30-12:30 Contact: Giandomenico Orlandi/ Giacomo Albi. Tue, 16 Dec 2025 08:30:00 +0100 https://www.di.univr.it/?ent=seminario&rss=0&id=6734 https://www.di.univr.it/?ent=seminario&rss=0&id=6856 Relatore: Erik Garrison; Provenienza: University of Tennessee; Data inizio: 2025-12-17; Ora inizio: 8.30; Note orario: Aula T06; Referente interno: Rosalba Giugno; Riassunto: Mini-course (6 hours) on quot;Algorithms to Build and Understand the Human Pangenomequot; within the course quot;Programming for Bioinformaticsquot; in the Master Programme in Medical Bioinformatics 17 Dicembre 2025 - Dalle 8:30 alle 11:30 - Aula T06 18 Dicembre 2025 - Dalle 10:30 alle 13:30 - Aula T06. Wed, 17 Dec 2025 08:30:00 +0100 https://www.di.univr.it/?ent=seminario&rss=0&id=6856