Alice Guionnet is a French mathematician and Research Director at the CNRS, affiliated with the Unité de Mathématiques Pures et Appliquées (UMPA) at École Normale Supérieure de Lyon since 2005. Her research focuses on random matrix theory, probability theory, and their applications in statistical mechanics and large deviations principles. She completed her doctorate in 1995 with a thesis on spin glass dynamics, supervised by Gérard Ben Arous. Her work includes groundbreaking contributions such as the Single Ring Theorem (2009) and studies on large deviations for eigenvalues of Wigner and heavy-tailed matrices. She has been honored with the Oberwolfach Prize (1998), Loève Prize (2009), and Blaise Pascal Medal (2018), and was elected to the French Academy of Sciences in 2017. Key research interests include non-linear Wigner spiked models, spectral phase transitions, and free probability. She has authored over 80 publications, including influential papers on matrix models, eigenvector delocalization, and stochastic processes in disordered systems. Her ERC Project LDRAM (Large Deviations in Random Matrices) explores asymptotic behaviors of random matrix ensembles. Education: PhD from École Normale Supérieure (1995), MSc in Mathematics (ENS Paris, 1989). Labs/Teams: UMPA Lyon, collaborations with CNRS and international institutions. Grants/Awards: Simons Investigator (2012–2015), CNRS Silver Medal (2010).
Sylvain Faisan is a permanent Assistant Professor at ICube - MIV (University of Strasbourg, France). His research focuses on image processing, statistical modeling, and geometry, with applications in medical imaging and neuroscience. He works on advanced methodologies integrating machine learning and mathematical frameworks. Key Research Areas: Polarimetric image processing, retinal image registration, 3D statistical model comparison, topology-preserving image deformation, and fMRI brain mapping Technical Expertise: Bayesian inference, non-local means filtering, reversible jump MCMC algorithms, causal modeling, and constrained optimization His publications demonstrate interdisciplinary applications in optics, biomedical imaging, and computational anatomy. He contributes to developing algorithms that maintain physical admissibility and topological integrity in complex imaging problems.
Jamal Atif is a Professor at Paris-Dauphine University and holds multiple significant leadership positions including Project Manager for 'Data Science and Artificial Intelligence' at the Institute of Information Sciences and their Interactions (INS2I) of the CNRS, Deputy Scientific Director of 3IA PRAIRIE, Head of the MILES team/project at LAMSADE (UMR CNRS-Université Paris-Dauphine), Co-leader of the Transverse Artificial Intelligence Program at PSL University, and Director of the Dauphine Numérique program. Professor Atif's primary research focuses on the foundations of responsible artificial intelligence, with specific expertise in privacy preservation in machine learning, robustness of deep learning algorithms to malicious attacks, causality, and explainability. His work bridges theoretical foundations with practical applications in security and reliability of AI systems. He has developed innovative approaches to address adversarial vulnerabilities in machine learning models and has made significant contributions to privacy-preserving techniques in data analysis. His publication record demonstrates a consistent focus on robust and trustworthy AI systems, with recent work exploring differential privacy in clustering, adversarial robustness, and explainable AI. The research spans theoretical foundations in logic and knowledge representation to practical applications in finance, healthcare, and computer vision. His publications appear in top-tier venues including Machine Learning journal, Neural Information Processing Systems, and International Joint Conferences on Artificial Intelligence. Scientific Awards: Recipient of two awards from the North American Society of Radiology for his thesis work Professor Atif has co-supervised or is currently supervising around fifteen doctoral students, demonstrating his commitment to mentoring the next generation of AI researchers. His leadership extends to directing major institutional programs including Dauphine Numérique and the Transverse Artificial Intelligence Program at PSL University, where he shapes strategic research directions in AI. He leads the MILES team/project at LAMSADE, which focuses on foundational aspects of machine learning and artificial intelligence. The team's research spans theoretical aspects of learning algorithms to practical applications requiring robust and reliable AI systems, with particular emphasis on security and privacy considerations in modern machine learning deployments.
Frédéric Pascal is a Full Professor at CentraleSupélec, part of the University of Paris-Saclay, and a member of the Laboratoire des Signaux et Systèmes (L2S). His research focuses on statistical signal processing, machine learning, and robust estimation techniques, with applications in radar detection, covariance matrix estimation, and information geometry. He has held roles including Coordinator of AI activities at CentraleSupélec and Head of the "Signals and Statistics" group at L2S. His academic journey includes a PhD from University Paris X – Nanterre (2006) and an HDR (2012) from University Paris-Sud. His work emphasizes adaptive signal processing in non-Gaussian environments, with contributions to robust covariance estimation, M-estimators, and applications in radar systems and biomedical signal processing. He has authored over 100 journal/conference papers and serves as an Associate Editor for IEEE Transactions on Signal Processing and Elsevier Signal Processing. Current research interests include AI transparency, data-driven methods for industry 4.0, and health-related signal analysis.
Ammar Mian is an Associate Professor at Université Savoie Mont Blanc, affiliated with the LISTIC lab and Polytech Annecy-Chambéry. He holds a PhD from CentraleSupélec (2016-2019) and conducted postdoctoral research at Aalto University (2019-2020). His research focuses on statistical signal processing, machine learning, and Riemannian geometry with applications in remote sensing and frugal computations. He leads the Qanat project, an experiment tracking tool for reproducible research. Research interests include covariance-based methods for SAR image analysis, robust detection algorithms for sonar and GPR systems, and optimization on Riemannian manifolds. His work emphasizes reproducibility in ML and efficient computational techniques for resource-constrained environments. Key contributions include real-time SAR time-series change detection, robust classification using second-order deep learning models, and novel methods for handling missing data in EEG signals. His recent articles (2023-2025) explore reproducibility frameworks, GPR-based object classification, and Riemannian geometry applications. No awards listed, but maintains active collaborations through LISTIC and industry partnerships. Advises students via internship programs (e.g., Federated ML energy cost analysis). Lab work involves developing open-source tools like Qanat for experiment management and reproducibility.
Mikael de la Salle is a CNRS Senior Researcher (Directeur de recherche) at the Institut Camille Jordan, Université Claude Bernard Lyon 1. He received his PhD from Université Paris 6 in 2009 and his Habilitation from ENS de Lyon in 2016. Previously, he held positions at Institut de Mathématiques de Jussieu, DMA, Laboratoire de Mathématiques de Besançon, and École Normale Supérieure de Lyon. During 2023-2024, he was a member at the Institute for Advanced Study in Princeton. His research focuses on interactions between functional analysis and group theory, including operator algebras, harmonic analysis, ergodic theory, and geometric group theory. His publications demonstrate expertise in non-commutative functional analysis, group representations, and geometric properties of discrete groups. He has received the Prix Charles-Louis de Saulce de Freycinet (2021) and was an invited speaker at the International Congress of Mathematicians (2022). He advises or has advised several students including Ignacio Vergara, Emilie Mai Elkiaer, Guillaume Dumas, and Martin Gilabert Vio. He organizes academic events such as the Entropies School/Workshop (2018) and serves on editorial boards for Confluentes Mathematici, Mathematische Zeitschrift, and Proceedings of the London Mathematical Society.
Sophie Huiberts is a CNRS researcher at LIMOS, Clermont Auvergne University in Clermont-Ferrand since fall 2023. Previously, she was a Simons Junior Fellow at Columbia University in New York City, hosted by Tim Roughgarden. She completed her PhD research at Centrum Wiskunde & Informatica in Amsterdam under Daniel Dadush and received her doctorate in 2022 from Utrecht University. Dr. Huiberts specializes in theoretical aspects of mathematical optimization, particularly focusing on the gap between practical performance and theoretical predictions of linear programming algorithms. Her research examines software implementations like Gurobi, CPLEX, SCIP, and HiGHS to understand why these algorithms perform better in practice than worst-case analysis would suggest. She has made significant contributions to smoothed analysis of the simplex method, establishing both upper and lower bounds on its complexity under perturbations of worst-case inputs. Analysis of her publication record shows consistent focus on bridging theoretical computer science with practical optimization methods. Her work spans linear programming theory, integer programming, combinatorial optimization, and computational geometry, with particular emphasis on understanding the geometric properties of optimization problems and the behavior of algorithms on real-world instances. Simons Junior Fellowship Dr. Huiberts maintains active engagement with the research community through social media platforms including Mastodon and Bluesky, and produces high-quality recordings of her research talks available on YouTube. She has made a conscious decision to stop air travel since 2023 due to climate concerns, demonstrating commitment to sustainable research practices while maintaining scientific connections through digital means. She is affiliated with LIMOS (Laboratoire d'Informatique, de Modélisation et d'Optimisation des Systèmes), a research laboratory at Clermont Auvergne University focused on computer science, modeling, and optimization systems, where she continues her investigations into the theoretical foundations of practical optimization algorithms.
Dr. Tim Seppelt is a postdoctoral researcher at the IT University of Copenhagen , working under the mentorship of Prof. Radu Curticapean. Previously, he earned his PhD from RWTH Aachen University with supervisors Prof. Martin Grohe and Prof. Michael Schaub. His research focuses on theoretical computer science, specifically homomorphism indistinguishability , a framework connecting graph isomorphism, quantum information, and logical equivalences. Current Role: Postdoc in Theoretical Computer Science, ITU Education: PhD in Computer Science, RWTH Aachen University Tim's work addresses algorithmic meta-theorems for homomorphism indistinguishability over minor-closed and treewidth-bounded graph classes. He has extended Lovász-type results to CMSO2 logic, resolved complexity conjectures for the Lasserre hierarchy, and classified quantum group-induced indistinguishability relations. His research spans quantum computing , graph algorithms , and descriptive complexity , often intersecting with applications in machine learning and finite model theory. Recent publications include a 2025 paper on quantum group-driven homomorphism indistinguishability and a 2024 journal article on logical equivalences and forbidden minors. He presented at workshops like the Graph Learning Meets TCS (Simons Institute, 2025) and delivered tutorials on finite model theory at Finite and Algorithmic Model Theory 2025 (Les Houches, France).
Sylvain Carrozza is a Teacher-Researcher (Maître de Conférences / Associate Professor) at the Institut de Mathématiques de Bourgogne (IMB), University of Burgundy, France. His work bridges gravitational physics, quantum theory, and combinatorics, with a focus on tensor models and group field theory. Education: PhD in Mathematical Physics (2013), Université Paris-Saclay (formerly Paris-Sud 11) Agrégation de Mathématiques (2010), École Normale Supérieure de Lyon Master in Theoretical Physics (2009), École Normale Supérieure de Lyon His research explores the large N limit of random tensors for quantum gravity in dimensions ≥3, non-perturbative quantum field theory , and boundary dynamics in generally covariant theories . Recent work extends these techniques to quantum information, particularly multipartite entanglement and quantum reference frames. Key trends in his publications include: Advancements in tensor models and group field theory for quantum gravity Analysis of edge modes and boundary effects in gauge and gravitational theories Applications to quantum information , such as error correction and entanglement Renormalization and scaling limits in multi-matrix and tensor systems Scientific Awards: Radboud Excellence Fellow (2020–2022) He is affiliated with the Mathematics-Physics team at IMB and contributes to the Parity BFC Mathematics Federation . His collaborations span institutions in France, Netherlands, Canada, and Germany.
Arshak Minasyan is an Associate Professor at CentraleSupélec – Université Paris-Saclay and a member of the Laboratoire Signaux et Systèmes (L2S), specifically the Modélisation et Estimation (GME) team. His research focuses on robust statistics and learning theory, particularly addressing computational and statistical complexities in high-dimensional data analysis. Education PhD in Mathematics from Yerevan State University (YSU), supervised by Prof. Victor Ohanian. Master’s in Mathematics from Skoltech. Bachelor’s in Mathematics from Higher School of Economics (HSE). Research Interests His work explores robust statistical methods for high-dimensional data, including outlier-robust algorithms, optimal estimation techniques, and the intersection of learning theory with statistical guarantees. Key areas include feature matching under noise, mean and covariance estimation, and generative modeling with distributional robustness. Key Contributions Notable contributions include the All-in-one Robust Estimator of the Gaussian Mean (Annals of Statistics, 2022) and outlier-tolerant methods for feature matching. His research emphasizes both theoretical foundations and computational tractability, with applications in signal processing and machine learning. Teaching Lecturer of “High-dimensional statistics” at Université Paris-Saclay. Teaching Assistant roles at CREST-ENSAE, HSE, and Skoltech in courses like Probability Theory, Stochastic Processes, and Biostatistics. Labs & Collaborations Affiliated with the L2S lab’s Modélisation et Estimation team, contributing to projects like COMEDY (Control of Dynamical Systems) and SYCOMORE (Signal Processing).
Jaouad Mourtada is an Assistant Professor in the Department of Statistics at ENSAE Paris, a founding member of the Institut Polytechnique de Paris, and a permanent member of CREST (Center for Research in Economics and Statistics). Since September 2020 he has held this faculty position, after completing a post-doctoral fellowship at the University of Genoa and earning his PhD from École Polytechnique. Education PhD in Statistics, École Polytechnique (2016–2019) MSc in Mathematics, "Probability and Random Models", Université Pierre et Marie Curie (2016) MSc in Mathematics, "Fundamental Mathematics", Université Pierre et Marie Curie (2015) BSc in Mathematics, Université Pierre et Marie Curie & École Normale Supérieure (2013) Student at École Normale Supérieure (2012–2016) Research Interests Mourtada’s work lies at the intersection of statistics and learning theory, with a focus on understanding the fundamental complexity of prediction and estimation tasks. His interests span: High-dimensional statistics and minimax theory Statistical learning theory and generalization bounds Online learning, regret minimization, and expert aggregation Density estimation and robust statistics Random forests, kernel methods, and convex optimization Research Output Trends Across more than fifteen recent publications, Mourtada has systematically advanced the understanding of statistical and computational limits in learning. His contributions range from exact minimax analyses of linear least squares and novel robust regression guarantees to refined PAC-Bayesian bounds for aggregation and sharp asymptotics for ridge regression. A recurrent theme is the development of estimators that achieve optimal or near-optimal rates while remaining computationally tractable and adaptive to unknown parameters. Scientific Awards & Recognition While the provided materials do not list specific awards, his sustained publication record in top venues such as Annals of Statistics , Journal of Machine Learning Research , Journal of the European Mathematical Society , and leading ML conferences (NeurIPS, COLT, AISTATS) attests to significant peer recognition. Teaching & Mentoring Mourtada has extensive teaching experience at both undergraduate and graduate levels, covering probability, statistics, and machine learning. Courses delivered include: Statistical Learning Theory (M2 Data Science, École polytechnique & ENSAE) Probability Theory (ENSAE) Python for Probability, Statistics, and Machine Learning (École polytechnique) Optimization for Data Science (M2 Data Science, École polytechnique) Laboratories & Collaborations He is affiliated with CREST/ENSAE and has previously collaborated with the Laboratory for Computational and Statistical Learning at the University of Genoa, the Center for Applied Mathematics (CMAP) at École Polytechnique, and maintains ongoing research ties with international scholars in statistical learning and optimization.
Roland Badeau is a Full Professor in the Signal, Statistics and Learning (S2A) team within the Image, Data, Signal (IDS) Department at Télécom Paris, Institut Polytechnique de Paris. His primary affiliation is with the Information Processing and Communication Laboratory (LTCI). Research Interests: Badeau specializes in statistical modeling of non-stationary signals, with core expertise in adaptive high-resolution spectral analysis and Bayesian extensions to Non-negative Matrix Factorization (NMF). His work spans room acoustics (stochastic reverberation models), data representation (dimensionality reduction, time-frequency analysis), probabilistic latent variable modeling, and algorithm development (Bayesian estimation, optimization methods, fast adaptive algorithms). Applications focus on audio/music processing including source separation, denoising, dereverberation, multipitch estimation, and automatic music transcription, with extensions to biomedical data analysis and digital communications. Key Trends in Publications: Recent work centers on Statistical Wave Field Theory, establishing mathematical frameworks for reverberation modeling using energy-stress tensor formalism and Riemannian geometry. His publications demonstrate a progression from foundational signal processing algorithms (e.g., YAST, ESPRIT) to physics-informed approaches for polyhedral rooms and frequency-dependent attenuation, with strong emphasis on Bayesian and alpha-stable distribution methods for robust audio separation. Academic Leadership: Badeau supervises doctoral and master’s theses while leading teaching units at Télécom Paris. He serves as the TSIA study track supervisor (Signal Processing for Artificial Intelligence) and Master ATIAM correspondent. His team (S2A) develops tools like DESAM for joint source separation and multi-track coding.
Philippe Sosoe is an Associate Professor in the Department of Mathematics at Cornell University, New York, and currently a Visiting Professor at the UMPA Laboratory (École Normale Supérieure de Lyon) from April 1 to April 30, 2025. His research focuses on stochastic methods applied to problems in mathematical physics, including Gibbs measures for dispersive PDEs, random matrix theory, and percolation models. He holds a PhD from Princeton University (2014) for his work on fluctuation bounds in disordered systems. During his visit to Lyon, he collaborates with Nikolay Tzvetkov and others on Gibbs measures for the nonlinear Schrödinger equation, dispersive PDEs with random data, and quasi-invariance of measures. He has received prestigious awards such as the NSF CAREER Award (2023–2027) and the Simons Collaboration Grant (2021–2026). Key research interests include the asymptotic behavior of dispersive PDE solutions, KPZ universality class phenomena, and sublinear variance in percolation models. His work bridges stochastic analysis, probability theory, and mathematical physics, often involving rigorous treatments of high-dimensional systems and random matrix ensembles. He is also engaged in academic activities such as lecturing on ergodicity of Dyson Brownian motion and Gibbs measures associated with nonlinear Schrödinger equations during his stay at ENS de Lyon.
Francesco Arzani is a junior professor at INRIA Paris and a member of the QAT team within the Department of Informatics at École Normale Supérieure. He obtained his PhD from ENS/PSL and Laboratoire Kastler Brossel, followed by post-doctoral appointments at LIP6, LORIA and Freie Universität Berlin. His research sits at the intersection of quantum optics and quantum information, with a focus on continuous-variable systems, bosonic error-correcting codes and fault-tolerant quantum computation. Education: PhD in Physics, École Normale Supérieure & Paris Sciences Lettres (2018) Graduate studies in quantum optics, Laboratoire Kastler Brossel Research interests: Arzani’s work targets the theory–experiment interface of continuous-variable quantum information processing. He designs non-Gaussian resources (photon subtraction, engineered squeezing), develops bosonic error-correcting codes (notably Gottesman–Kitaev–Preskill codes) and investigates fault-tolerant architectures that remain realistic with present-day photonic technology. Key themes include universal CV gate sets, entanglement in optical frequency combs, measurement-based quantum computation and the mathematical structure of lattice codes in phase space. Publication trends: Across 20+ peer-reviewed articles and several preprints, Arzani consistently advances from abstract code design to experimental feasibility . Recent works (2022-2025) emphasize fault-tolerant thresholds and symmetry-enhanced variational algorithms, while earlier papers map out practical methods to generate and certify multimode entanglement in femtosecond-frequency-comb platforms. Scientific outreach & awards: He has delivered invited seminars at Xanadu, WACQT, NC State and numerous international workshops; serves as referee for leading quantum journals; and maintains open-source slides and posters for community use. No competitive fellowships or prizes are explicitly listed in the supplied sources. Team & grants: Arzani leads independent research funded through the INRIA Junior Professor Chair scheme and participates in the French “Défi EQIP” quantum initiative. He collaborates closely with experimental groups at Kastler Brossel, Xanadu and within the EU continuous-variable quantum community, co-supervising graduate students and post-docs in the QAT team.
Djalil CHAFAÏ is a University Professor of Mathematics at Université Paris-Dauphine - PSL, with dual affiliation at CEREMADE (Centre de Recherche en Mathématiques de la Décision) and DMA (Département de Mathématiques et Applications) at École normale supérieure (Paris) - PSL. He currently serves as Directeur des études du DMA (2021-2026) and Directeur scientifique du RNBM (2021-2025). His extensive research spans multiple areas of probability theory, mathematical physics, and applied mathematics. CHAFAÏ's research interests center around geometric and probabilistic functional analysis, random matrices, random graphs, free probability, and high-dimensional phenomena. His work connects mathematical theory with applications in biology, physics, and data science. He has made significant contributions to understanding cutoff phenomena in high-dimensional diffusions, Riesz energy problems, and Coulomb gases. His research often combines theoretical analysis with visual illustrations created using computational tools like Octave, Python, and Julia. Analysis of his recent publications reveals a strong focus on cutoff phenomena in various stochastic processes, equilibrium measures in potential theory, and the mathematical properties of random matrix ensembles. His work demonstrates a consistent pattern of bridging abstract mathematical concepts with concrete physical phenomena, particularly in statistical physics. The interdisciplinary nature of his research is evident in the diverse applications ranging from mathematical biology to data science. CHAFAÏ has successfully advised numerous doctoral students whose work continues to influence the field. His current doctoral students include Samuel Chan-ashing, Rémi Bonnin, and Kewei Pan, while his former students have gone on to positions at prestigious institutions worldwide. He is actively involved in the mathematical community through organizing conferences, seminars, and workshops, including the Matrices Et Graphes Aléatoires (MEGA) project and the Conviviality project funded by ANR.
