5 décembre 2024 @ 14h00 – 15h00 – Speaker: Paul Catala Title: An approximate joint diagonalisation algorithm for algebraic recovery from moments Location: CRAN’s Room at Henri Poincaré Building (4th Floor), Faculté des Sciences et Technologies Abstract: Many problems in imaging and data science require to reconstruct, from partial observations, highly concentrated signals, such as pointwise sources or contour lines. This work introduces […]
10 janvier 2025 @ 9h00 – 12h00 –
16 janvier 2025 @ 13h00 – 14h30 –
25 février 2025 @ 14h00 – 15h00 – Speaker: Guilherme Holsbach Costa (University of Caxias do Sul), visiting professor Title: On the Super-Resolution Image Reconstruction and Phase Correlation Image Registration: methods, insights, and challenges Abstract: In this seminar, we will explore classical methods in image processing, focusing on two key topics. First, we will discuss the methodology of analytical approaches applied to the […]
6 mars 2025 @ 13h00 – 14h30 –
25 avril 2025 @ 13h00 – 14h30 –
22 mai 2025 @ 13h00 – 14h30 –
19 juin 2025 @ 9h00 – 12h00 –
3 juillet 2025 @ 9h45 – 10h45 – Speaker: Vahid Shahverdi (KTH, Sweden), https://sites.google.com/view/vahidshahverdi/ Title: Neuroalgebraic Geometry: An Algebraic Perspective on Neural Networks Abstract: In this talk, I introduce neuroalgebraic geometry: a unified algebraic-geometric framework for analyzing neural networks via polynomial activations. Our central object is the neuromanifold, the function space realized by a network, which forms a semi-algebraic variety when activations are polynomial. […]
3 juillet 2025 @ 10h45 – 11h30 – Speaker: Clara Dérand (CRAN) Title: Identifiability of Deep Polynomial Neural Networks Abstract: Polynomial Neural Networks (PNNs) possess a rich algebraic and geometric structure. However, their identifiability — a key property for ensuring interpretability — remains poorly understood. In this work, we present a comprehensive analysis of the identifiability of deep PNNs, including architectures with and without […]
