Видео

MaLGa Colloquium, October 23rd, 16:00 Speaker: Max Welling, University of Amsterdam Title: Physics Priors for Machine Learning and Machine Learning to Solve Physics Abstract: Good neural architectures are rooted in good inductive biases (a.k.a. priors). Equivariance under symmetries is a prime example of a successful physics inspired prior which sometimes dramatically reduces the number of exam...

MaLGa Colloquium, 26 May 2023 Speaker: Alfio Quarteroni Affiliation: Politecnico di Milano and EPFL Title: Physics-based and data-driven mathematical models for the simulation of the heart function Abstract: This presentation will focus on an integrated numerical model to simulate the cardiac function. Physics-based models will represent the backbone of our approach, however their synergistic u...

MaLGa Colloquium - May 9th 2023 Speaker: Ben Recht, University of California, Berkeley Title: Cancer Screening: A Parable of Prediction Abstract: The central promise of machine learning in medicine is improving predictions in healthcare settings. But this begs the question of the importance of prediction in medicine. This talk will examine the value of prediction in healthcare through the histo...

MaLGa Colloquium - 17 April 2023 Speaker: Nelly Pustelnik, ENS Lyon Title: Fast unrolled proximal algorithms to design stable and efficient neural network architectures Abstract: Deep learning has revolutionized many image processing tasks including inverse problems. However, it remains a gap between the black-box deep learning approaches and the more recently unrolled deep learning techniques ...

MaLGa Colloquium 3rd April 2023 Speaker: Sayan Mukherjee, MPI Leipzig Title: Modeling shapes and surfaces - Geometry meets machine learning Abstract: We will consider modeling shapes and fields via topological and lifted-topological transforms. Specifically, we show how the Euler Characteristic Transform and the Lifted Euler Characteristic Transform can be used in practice for statistical analy...

MaLGa Colloquium, 20th March 2023 Speaker: Lamberto Ballan, University of Padova Title: Context-aware motion prediction and embodied social navigation Abstract: A key goal of computer vision is to understand complex visual scenes, by recognizing visual concepts, localizing them, and understanding their interactions within the scene. However, most of the recent successes in visual recognition ar...

MaLGa Colloquia series 20th February 2023 Speaker: Antonio Celani, Senior Researcher at ICTP Title: The uses of memory in olfactory search Abstract: Many organisms, from insects to mammals, have developed exquisite skills in searching for sources of odor, from huge distances, and in turbulent atmospheric conditions. How are sequences of very sparse odor detections encoded in the memory of the s...

MaLGa Colloquia series 6th March 2023 Speaker: Peter Dayan, Director at the Max Planck Institute for Biological Cybernetics in Tübingen Title: Risk, Replay and Rehearsal Abstract: Risk, which arises when outcomes are not completely certain, occupies a central role in both the theory and practice of decision-making, and is especially important in the context of survival. Fortune famously favours...

MaLGa Seminar - Analysis and Learning series Speaker: Romain Petit Affiliation: UniGe | MaLGa Date: Monday January 30th, 2022 Title: Grid-free sparse regularization Abstract: The recovery of a sparse vector from noisy linear measurements has found numerous applications in imaging and statistics / machine learning. A popular method for performing this reconstruction is (discrete) l1 regularizati...

MaLGa Seminar, Series - Analysis and Learning This event is part of the Ellis Genoa activities. Speaker: Elena Beretta Affiliation: New York University Abu Dhabi Date: Monday July 18th, 2022 Title: Mortality containment vs. economics opening: optimal policies in a SEIARD model Abstract: We extend the classic approach SIR to a SEAIRD model with policy controls. A social planner’s objective refle...

MaLGa Seminar Series - Machine Learning and Vision This event is part of the Ellis Genoa activities. Speaker: Matteo Bianchi Affiliation: Centro E. Piaggio, University of Pisa Date: Monday June 6th, 2022 Title: Model-based and data-driven approaches for modelling human sensory-motor system: applications to advanced sensing systems and autonomous robotic grasping Abstract: Human example represen...

MaLGa Seminar Series - Analysis and Learning. This event is part of the Ellis Genoa activities Speaker: Tatiana Bubba Affiliation: University of Bath Date: Monday June 27th, 2022 Title: Learning a microlocal prior for limited angle tomography Abstract: Limited-angle tomography is a highly ill-posed linear inverse problem It arises in many applications, such as digital breast tomosynthesis. Reco...

MaLGa Seminar Series - Statistical Learning and Optimization. This event is part of the Ellis Genoa activities. Speaker: Johannes Maly Affiliation: Catholic University of Eichstaett-Ingolstadt Date: Monday May 30th, 2022 Time: 16:00 p.m. Title: The implicit bias of gradient descent and its applications Abstract: In deep learning it is common to overparameterize neural networks, that is, to use ...

MaLGa Seminar Series - Analysis & Learning. This event is part of the Ellis Genoa activities. Speaker: Johannes Hertrich Affiliation: Technische Universität Berlin Date: Monday, May 9th, 2022 Title: Stochastic Normalizing Flows for Inverse Problems: a Markov Chains Viewpoint Abstract: Normalizing flows aim to learn the underlying probability distribution of given samples. For this, we train a d...

Michael Unser (EPFL) - The Radon transform, neural networks and splines

Patrizio Frosini (University of Bologna) - Group equivariant non-expansive operators

Donatello Conte (Université de Tours) - Structural Pattern Recognition meets Computer Vision

Tomaso Poggio (MIT) - Thoughts on today’s learning theory

Andrea Manzoni (POLIMI) - Deep learning reduced order models for numerical approximation of PDEs

Marcello Carioni (University of Cambridge) - Generalized conditional gradient methods

Massimo Vergassola (CNRS) - Reinforcement learning for animal behavior

Claire Vernade (Deepmind) - Non-Stationary Delayed Bandits with Intermediate Observations

Massimo Fornasier (TUM) - Three Mathematical Tales of Machine Learning

Gian Maria Marconi (RIKEN) - Structured Prediction for CRiSP Inverse Kinematics Learning

Maria Pia Sormani (University of Genoa)

Matteo Monti - Unitarization of the Radon transform on homogeneous trees