Find Academic Supervisors

Francesco RussoView profile

Professor

• Stochastic Analysis
• Financial Mathematics
• Probabilistic Models in Mathematical Physics

Francesco Russo is a Professor of Exceptional Class at ENSTA Paris under the Applied Mathematics Unit (UMA) . He has held academic positions at INRIA-Ecole des Ponts (2008-2010) and Paris 13 University (1994-2008) , where he led the Probability and Statistics Team and the Financial Engineering Option in the MACS course. His research spans Stochastic Analysis , Financial Mathematics , and Probabilistic Models in Mathematical Physics , with applications to energy systems, control theory, and nonlinear PDEs. He co-organizes international seminars and conferences, including the Seminar in Probability-Statistics-Control and the Day Around Stochastic PDEs . Research Themes : Stochastic calculus via regularization, path-dependent PDEs, BSDEs, non-semimartingale models, fractional Brownian motion, and McKean-Vlasov equations with irregular coefficients. Projects : Leads the SDAIM (2023-27) project funded by ANR (France) and FAPESP (Brazil). Coordinated the ANR MASTERIE (2011-2013) program. Teaching : Courses include Elementary Stochastic Calculus (ENSTA), Discrete Models in Finance (ENSTA), and Stochastic Calculus (Master Paris-Saclay). Collaborations : Organizes seminars with institutions such as Luiss University (Rome) and EPFL (Lausanne). Collaborates with Brazilian teams (UNICAMP) and French institutions (CMAP, CentraleSupélec).

Youssef MarzoukView profile

Professor

• Uncertainty Quantification
• Bayesian Computation
• Inverse Problems
• +4 more

Youssef Marzouk is a Professor of Aeronautics and Astronautics at MIT, serving as co-director of the MIT Center for Computational Engineering and director of the Aerospace Computational Design Laboratory. His research focuses on integrating physical modeling with statistical inference, emphasizing Bayesian computation, uncertainty quantification, and optimal experimental design. He holds a SB, SM, and PhD from MIT and has been recognized with prestigious awards including the DOE Early Career Award and the Junior Bose Teaching Prize. Education: PhD in Aeronautics and Astronautics, MIT SM in Aeronautics and Astronautics, MIT SB in Aeronautics and Astronautics, MIT Research Interests: Uncertainty Quantification techniques for complex systems Bayesian computational methods and inverse problem solutions Optimal experimental design strategies Interdisciplinary applications in geophysics, environmental science, and engineering Awards: 2022: Report to the President, Center for Computational Science and Engineering 2021: Bayesian Inference Software Framework (hIPPYlib-MUQ) 2012: MIT School of Engineering Junior Bose Award 2010: DOE Early Career Research Award Labs & Leadership: Aerospace Computational Design Laboratory (Director) MIT Center for Computational Engineering (Co-Director) Editorial Board roles: SIAM Journal on Scientific Computing, Advances in Computational Mathematics

Hugo Georges Victor LavenantView profile

Assistant Professor

• Optimal Transport
• Calculus of Variations
• Convex Analysis
• +3 more

Hugo Georges Victor Lavenant serves as Assistant Professor in the Department of Decision Sciences at Bocconi University, Milan, where he has held a faculty position since 2020. Previously, he completed a postdoctoral fellowship at the University of British Columbia (2019-2020) under the Pacific Institute of Mathematical Sciences and earned his PhD in Mathematics from Université Paris-Sud (2016-2019) under Filippo Santambrogio's supervision. His academic foundation includes: PhD in Mathematics, Université Paris-Sud (2016-2019) Studies at École Normale Supérieure (2012-2016) covering mathematics, physics, history, and philosophy of science Classes préparatoires in mathematics and physics (2010-2012) Lavenant's research centers on optimal transport theory and its applications across mathematical disciplines. He investigates geometric structures in Wasserstein spaces, develops numerical methods for dynamical optimal transport, and bridges theoretical advances with Bayesian statistics. His work demonstrates particular innovation in trajectory inference for biological data and dependence measures for random measures, connecting pure mathematics with computational statistics. Recent publications reveal accelerating interdisciplinary impact, with 2024-2025 works extending optimal transport to machine learning (kernel methods, variational inference) and data science (opinion dynamics, single-cell analysis). This trajectory shows increasing methodological sophistication in handling measure-valued mappings and non-smooth geometries while maintaining computational tractability. Award recognition includes: Pacific Institute of Mathematical Sciences Postdoctoral Fellowship Lavenant actively mentors early-career researchers through formal advising relationships and collaborative projects. He currently supervises two PhD candidates (George Kanchaveli and Francesco Mascari, co-advised with Marta Catalano) and has guided Master's students including Mathis Hardion and Niccolò Bargellini. His teaching portfolio spans advanced analysis, optimization, and real analysis courses at Bocconi, reflecting his commitment to mathematical rigor in education. He operates within Bocconi's Decision Sciences ecosystem while maintaining international collaborations with researchers at UBC, Université Paris-Sud, and statistical groups worldwide. Current projects focus on entropy-based transport methods and geometric approaches to nonparametric statistics, positioning his work at the intersection of theoretical mathematics and data-driven applications.

SHEN LeiView profile

Researcher

• Multiscale Modeling and Simulation
• Materials Informatics
• Computational Physics
• +3 more

SHEN Lei is a researcher at the National University of Singapore (NUS), affiliated with the Department of Physics. With a PhD in Physics from NUS, he specializes in Multiscale Modeling and Simulation and Materials Informatics , leveraging machine learning and computational methods for advanced materials discovery. Research Focus: Density functional theory, molecular dynamics, finite element analysis, and data-driven design of materials. Teaching: Modules include Mechanics and Waves (PC1433), Applied Quantum Mechanics (PC2130B), and Mechanical Properties of Materials (ESP2109). His work spans spintronics, ferroelectricity, and energy storage materials, with recent publications on interatomic potentials, sliding heterostructures, and battery anodes. He has received the Teaching Commendation Award and declined the Lee Kuan Yew Postdoctoral Fellowship . Notable Trends: Recent articles emphasize machine learning in materials science, van der Waals heterostructures, quantum transport, and medical image analysis. Subfields include Rashba spin-orbit coupling, piezoelectric tensor modeling, and defect-informed neural networks. Scientific Awards: Teaching Commendation Award (AY15/16; AY16/17) Lee Kuan Yew Postdoctoral Fellowship (2014) (declined)

Rémi FlamaryView profile

Professor

• Machine Learning
• Optimal Transport
• Domain Adaptation
• +7 more

Rémi Flamary is a Professor in the Applied Mathematics department at École Polytechnique, France, and a member of the CMAP Laboratory. Previously, he held an Associate Professor position at Université Côte d'Azur in the Department of Electronics and Lagrange Laboratory. He completed his PhD at Rouen University under Alain Rakotomamonjy, focusing on statistical signal processing and optimization. His research interests span machine learning, optimal transport, domain adaptation, and their applications in biomedical engineering, energy, and remote sensing. He leads the development of the POT (Python Optimal Transport) library and contributes to projects like SKADA for domain adaptation. Education: PhD in Applied Mathematics, Rouen University (LITIS Laboratory) Research Focus: Flamary's work emphasizes leveraging optimal transport theory for machine learning tasks such as graph prediction, signal normalization, and cross-domain adaptation. His contributions include novel algorithms for unbalanced transport, semi-relaxed Gromov-Wasserstein distances, and end-to-end graph generation frameworks. He actively collaborates on applications in neuroscience, astronomy, and energy systems. Professional Activities: He has presented at NeurIPS and other top conferences, supervised PhD students like Cédric Vincent-Cuaz, and contributed to open-source software. His teaching includes courses on signal processing and machine learning at École Polytechnique.

• Discrete Mathematics
• Probability
• Theoretical Computer Science
• +6 more

Prasad Tetali is a Regents' Professor at the Georgia Institute of Technology , with appointments in both the School of Mathematics and the School of Computer Science . He also holds an adjunct professor position at Emory University's Mathematics and Computer Science departments. Education: PhD in Mathematics (1991) from the Courant Institute of Mathematical Sciences at NYU; MS in Mathematics (1987) from the Indian Institute of Science; Postdoc at AT&T Bell Labs His research spans Discrete Mathematics, Probability Theory, and Theoretical Computer Science , focusing on Markov chains, Isoperimetry, Combinatorics, Computational number theory, and Algorithms. Recent work includes applications to statistical physics models and hypergraph structures. He has served as Director of the ACO Ph.D. Program since 2019 and held leadership roles like Interim Chair of the School of Mathematics. His publications reflect a blend of discrete geometry, stochastic processes, and algorithmic analysis . Key Honors: AMS Fellow (2012) SIAM Fellow (2009)

Stefano MartinianiView profile

Assistant Professor

• Computational and Statistical Physics
• Materials Science
• Artificial Intelligence for Science
• +5 more

Stefano Martiniani is an Assistant Professor of Physics, Chemistry, Mathematics, and Neuroscience at New York University, affiliated with the Center for Soft Matter Research and the Simons Center for Computational Physical Chemistry. His interdisciplinary research explores computational physics of complex systems, including neural circuit theories, non-equilibrium statistical mechanics, and AI-driven materials discovery. He has pioneered methods for analyzing high-dimensional energy landscapes and received prestigious awards like the NSF CAREER Award (2024) and IUPAP Early Career Prize (2023). Education: PhD in Physics (2017), University of Cambridge MPhil in Physics (2013), University of Cambridge BSc in Physics (2012), Imperial College London Research Interests: His work bridges statistical physics and artificial intelligence, focusing on: Engineering disordered materials with tailored spectral properties Quantifying entropy production in active matter Developing open science frameworks like ColabFit for machine learning interatomic potentials Neural circuit models for cortical communication Grants & Collaborations: Funded by NSF, NIH, Chan Zuckerberg Initiative, and Simons Foundation. Leads interdisciplinary teams in computational physics, AI, and materials science. Labs/Initiatives: Core member of NYU's Center for Soft Matter Research; develops software tools like FReSCo and KLIFF-Torch for computational materials science.

Shahin ShahrampourView profile

Assistant Professor

• Machine Learning
• Optimization and Control
• Distributed Learning
• +5 more

Shahin Shahrampour is an Assistant Professor in the Department of Mechanical & Industrial Engineering at Northeastern University, with a courtesy appointment in Electrical & Computer Engineering. He holds a Ph.D. in Electrical & Systems Engineering from the University of Pennsylvania, an M.A. in Statistics from The Wharton School, and a B.S. in Electrical Engineering from Sharif University of Technology. Prior to Northeastern, he was at Texas A&M University and held a postdoctoral fellowship at Harvard University. Research Interests: Machine learning, optimization and control, distributed and sequential learning, with a focus on computationally efficient methods for data analytics. Specific areas include manifold optimization, online and reinforcement learning, non-convex optimization, and statistical signal processing. His work bridges theory and applications in networked systems and dynamic environments. Awards & Grants: Received a $500,000 NSF grant (2023) for distributed optimization in non-convex environments. Best Paper Award at IEEE ICASSP 2022 for contributions to signal processing. TEES Engineering Genesis Award (2020) for multidisciplinary research at Texas A&M. Principal Investigator on NSF-funded projects in collaborative online optimization and decentralized learning. Labs & Teams: Leads research in machine learning and control systems at Northeastern University, focusing on interdisciplinary challenges in distributed optimization and real-time learning. His lab emphasizes theoretical foundations and practical implementations in networked and dynamic systems.

• Reinforcement Learning
• Robotics
• Deep Learning
• +3 more

Glen Berseth is an Associate Professor in the Department of Computer Science and Operations Research at the University of Montreal and a Senior Academic Fellow at Mila – Quebec Institute for Artificial Intelligence. He is also a Canada CIFAR Chair in AI and Co-Director of the Montreal Robotics and Integrative AI Laboratory (REAL). His work focuses on reinforcement learning, robotics, and deep learning applied to autonomous systems. He holds a postdoctoral background from Berkeley Artificial Intelligence Research (BAIR), working under Sergey Levine. His research emphasizes real-world applications, including human-robot collaboration, continual learning, and multi-agent systems. He teaches courses on robot learning at the University of Montreal and Mila, covering cutting-edge techniques for general-purpose robots. Key research interests include reinforcement learning for robotics, adaptive interfaces, and sim-to-real transfer. His recent work addresses challenges in autonomous learning systems, such as robust locomotion control and efficient exploration strategies. Notable awards include the Canada CIFAR AI Chair. He has supervised numerous students, including PhD candidates Ozgur Aslan and Siddarth Venkatraman, and Master’s students like Roger Creus-Castanyer and Léa Demeule, focusing on topics like reinforcement learning and robotic control. Berseth leads research projects funded by organizations like the CRSNG, FCI, and MITACS, addressing topics such as modular lifelong learning and generalization in robotics. His lab, REAL, explores embodied AI and robotics integration.

Michael FriedlanderView profile

Professor

• Convex Optimization
• Machine Learning
• First-Order Methods
• +3 more

Michael Friedlander is a Professor of Computational Mathematics at the University of British Columbia (UBC), holding joint appointments in the Departments of Computer Science and Mathematics. He serves as Director of the UBC Institute of Applied Mathematics and is a Sauder School Distinguished Scholar. His research focuses on optimization theory, algorithms, and applications, particularly in convex optimization, first-order methods, and machine learning. Friedlander is a SIAM Fellow (2024) and has received multiple teaching awards, including the 2013 UBC CS Teaching Award. Education: PhD (2002) in Operations Research from Stanford University; BA (1993) in Physics from Cornell University. Research Interests: Friedlander’s work spans computational optimization, including first-order methods, gauge duality, federated learning, and applications in signal processing and quantum computing. His contributions include SPGL1, a widely used convex optimization solver, and foundational work on atomic decomposition and polar alignment in structured optimization. Key Contributions: His research bridges optimization with practical applications, such as compressed sensing and data science. Notable areas include dual methods for federated learning, low-rank spectral optimization, and algorithms for sparse recovery. He has advised numerous PhD students, including work on topics like gauge duality and federated learning contribution valuation. Awards & Honors: SIAM Fellow (2024) 2013 UBC CS Department Teaching Award CACS Outstanding Young Researcher Prize AI & Statistics 2009 Best Paper Award Advising & Grants: Supervised over 20 graduate students and postdocs. Active in grant-funded research, including projects on optimization algorithms and their industrial applications. Collaborates with institutions like Pacific Institute for Mathematical Sciences (PIMS) and the Data Science Institute at UBC. Labs & Affiliations: Affiliated with the Institute for Computing, Information and Cognitive Systems (ICICS), Data Science Institute, and Quantum Computing Research Cluster at UBC. Leads interdisciplinary efforts in applied mathematics and computational science.

Ali Ramezani-KebryaView profile

Associate Professor

• Machine Learning
• Optimization
• Privacy/security
• +3 more

Ali Ramezani-Kebrya is currently an Associate Professor (with tenure) in Computer Science at the University of Oslo. Previously, he was a postdoctoral fellow at the Laboratory for Information and Inference Systems (LIONS) at EPFL and the Vector Institute in Canada. His research focuses on large-scale and distributed machine learning, optimization, privacy/security, reinforcement learning, and communication/networking aspects of machine learning algorithms. He holds a Ph.D. from the University of Toronto. Key research interests include developing robust federated learning frameworks, addressing label shift in distributed systems, and improving the generalization capabilities of stochastic gradient descent methods. His work bridges theoretical foundations with practical applications in distributed optimization and privacy-preserving machine learning. Recipient of the NSERC Postdoctoral Fellowship (equivalent to NSF fellowship in the US) His publications emphasize advancements in distributed learning systems, robust optimization techniques, and theoretical guarantees for federated learning under covariate shifts. Current research trends explore Nash equilibrium-based approaches for robustness and communication-efficient algorithms in distributed settings.

James MurphyView profile

Associate Professor

• Machine Learning
• Harmonic Analysis
• Statistical Learning
• +5 more

James Murphy is an Associate Professor of Mathematics at Tufts University, with affiliate appointments in Computer Science and Electrical and Computer Engineering. His research focuses on the mathematics of artificial intelligence, statistical and machine learning, high-dimensional probability, and their applications to image/signal processing, molecular dynamics, and computational biology. He holds grants from the NSF, NGA, Camille & Henry Dreyfus Foundation, and others. His work integrates theoretical mathematics with computational tools, emphasizing open-source software development. Education: PhD in Mathematics (2015) and MS (2013) from University of Maryland, BS in Mathematics (2011) from University of Chicago. Research interests include machine learning theory (e.g., spectral clustering, unsupervised learning), harmonic analysis, graph theory, and applications to hyperspectral imaging, medical data analysis, and remote sensing. Notable projects involve Wasserstein-based methods for data clustering, diffusion geometry for image processing, and algorithms for anomaly detection in high-dimensional datasets. His research group, HILDA, collaborates on projects like GLIDE (biological network prediction), Fermat distances for clustering, and entropy-regularized optimal transport. He teaches courses in functional analysis and has contributed to open-access educational materials for calculus and algebra.

Samantha PettiView profile

Assistant Professor

• Bioinformatics
• Computational Biology
• Machine Learning
• +4 more

Samantha Petti is an Assistant Professor in both the Department of Mathematics (School of Arts and Sciences) and the Department of Computer Science (School of Engineering) at Tufts University. She is based at 177 College Avenue, Medford, MA. Her research focuses on computational biology, bioinformatics, and machine learning, with particular emphasis on protein structure analysis, genotype-phenotype mapping, and algorithm design for biological sequence analysis. She teaches courses such as Master's Thesis supervision, PhD Thesis guidance, and specialized topics in mathematics and computer science. Her work integrates interdisciplinary approaches, combining statistical methods, deep learning, and probabilistic models to address challenges in genomics, structural biology, and network science. Recent projects include developing end-to-end protein alignment tools and exploring sparse graph models for biological systems. Teaching: Supervises graduate thesis work (Master’s/PhD) and advanced courses in mathematics and computer science at Tufts. Lab/Team: Engages in collaborative research at the intersection of computational methods and biological systems.

Marco GrangettoView profile

Professor

• Computer Vision
• Image Processing
• Artificial Intelligence
• +5 more

Marco Grangetto serves as Full Professor in the Department of Computer Science at the University of Turin, coordinating research in image processing and computer vision. His expertise spans wavelets, image/video coding, data compression, error resilient video coding, and biomedical image processing, with significant contributions to ISO JPEG2000 standardization and editorial roles in IEEE Transactions on Multimedia. His educational background includes a PhD in Electrical and Communications Engineering (2003) and MSc in Telecommunications Engineering (1999), both from Politecnico di Torino. His research integrates Artificial Intelligence and Deep Learning with medical imaging and fundamental compression theory , producing innovations in neural network pruning, capsule networks, and entropy-based models. Recent work focuses on Covid-19 diagnosis from chest X-rays and efficient 3D scene modeling. His publication trends reveal dual trajectories: applied medical AI (Covid-19 diagnostics, lung cancer segmentation) and theoretical advances (learned compression, contrastive learning, bias mitigation). This bridges clinical validation with information-theoretic foundations, particularly in resource-constrained environments. Scientific recognition includes: Premio Optime by Unione Industriale di Torino (2000) Fulbright Grant for research at UC San Diego (2001) He maintains leadership through IEEE editorial positions, ISO standardization participation, and MPAI membership. His research group develops medical datasets (UniToChest, UniToPatho) while advancing neural network efficiency for clinical deployment. Current projects focus on entropy minimization techniques and unbiased representation learning for healthcare applications.

• Visual Media Compression
• Machine Learning
• End-to-End Optimization
• +5 more

Jona Ballé is an Associate Professor in the Department of Electrical and Computer Engineering at New York University Tandon School of Engineering. His research focuses on visual media compression, leveraging machine learning and end-to-end optimization to advance compression techniques for traditional and emerging modalities (e.g., AR, plenoptic imaging). He holds a PhD in signal processing from RWTH Aachen University (2012), followed by postdoctoral work at NYU’s Center for Neural Science and a Research Scientist role at Google (2017–2024). His contributions include foundational work on JPEG AI and leadership in conferences such as CLIC and DCC. Education: PhD in Signal Processing, RWTH Aachen University (2012) Master’s in Signal Processing, RWTH Aachen University (2007) Research: Ballé bridges machine learning and perceptual science to improve compression efficiency. Key areas include perceptual metrics, end-to-end optimization, and distributed coding. His work on Wasserstein distortion and Fourier basis models reflects cutting-edge advancements in compression theory. Impact: As co-organizer of the Challenge on Learned Image Compression (CLIC) and program committee member of the Data Compression Conference (DCC), he drives community progress. His industry collaborations and leadership in the JPEG AI standard highlight his translational research.