Find Academic Supervisors

Keith LevinView profile

Assistant Professor

• Network analysis
• Dimension reduction
• Concentration inequalities
• +5 more

Keith Levin is an Assistant Professor in the Department of Statistics at the University of Wisconsin-Madison, part of the School of Computer, Data & Information Sciences. He holds a Ph.D. in Computer Science from Johns Hopkins University (2017) and completed a postdoctoral fellowship at the University of Michigan. His research focuses on network analysis, dimension reduction, and statistical inference in complex systems, with applications to neuroscience and speech processing. Levin teaches courses on computational tools for data science and has developed methodologies for network bootstrapping, latent space modeling, and spectral graph analysis. His work bridges theoretical statistics with practical machine learning challenges, particularly in high-dimensional data settings. Research Interests: - Network analysis and community detection - Dimensionality reduction techniques - Latent space models and graph embeddings - Applications in neuroscience and speech processing - Statistical methods for large-scale data Selected Contributions: - Developed bootstrap methods for network data analysis - Advanced spectral embedding techniques for graph inference - Authored tools for clustering and motif detection in stochastic blockmodels

Christopher MuscoView profile

Assistant Professor

• Algorithmic Foundations of Data Science
• Machine Learning
• Numerical Linear Algebra
• +5 more

Christopher Musco is an Assistant Professor in the Department of Computer Science and Engineering at New York University’s Tandon School of Engineering. His research focuses on the algorithmic foundations of data science and machine learning, integrating theoretical computer science, numerical linear algebra, and optimization. He holds a Ph.D. from MIT and B.S. degrees in Applied Mathematics and Computer Science from Yale University. Education: Ph.D. in Computer Science, MIT B.S. in Applied Mathematics & Computer Science, Yale University Research Interests: Scalable machine learning, numerical linear algebra, randomized algorithms, sketching/streaming methods, and matrix approximation. His work emphasizes efficient processing of large datasets through algorithmic innovation and theoretical analysis. Publications: Over 50 peer-reviewed articles, including contributions to NeurIPS, SODA, COLT, and ICML, focusing on matrix approximation, spectral methods, and graph algorithms. Recent work includes breakthroughs in hierarchical matrix approximation and provably accurate nearest neighbor search. Awards & Funding: NSF CAREER Award Google Research Scholar Award Funding from NSF, DOE, and NYU grants Labs & Groups: Member of NYU’s Theoretical Computer Science Group and Visualization Imaging and Data Analysis Center (VIDA). Active in organizing the CS Theory Seminar and mentoring over 15 Ph.D. students and postdocs.

Madeleine UdellView profile

Assistant Professor

• Optimization
• Machine Learning
• Data Analysis
• +4 more

Madeleine Udell is an Assistant Professor of Management Science and Engineering at Stanford University, with affiliations at the Institute for Computational and Mathematical Engineering (ICME) and a courtesy appointment in Electrical Engineering. She holds a PhD from Stanford University in Computational and Mathematical Engineering and completed a postdoc at Caltech. Her research focuses on accelerating optimization and machine learning, with applications in healthcare, finance, and engineering. Udell's work bridges algorithmic efficiency, interpretability, and practical deployment in data-driven systems. Education: BS in Mathematics and Physics, Yale University (2009) PhD in Computational and Mathematical Engineering, Stanford University (2015) Research Interests: Her research emphasizes optimization algorithms, scalable machine learning techniques, and their applications in real-world domains. Key areas include low-rank matrix methods, automated machine learning (AutoML), and interpretable models for healthcare. She develops tools like OptiMUS for natural language-based optimization modeling and explores second-order methods for improving deep learning efficiency. Awards: Kavli Fellowship (2023) Alfred P. Sloan Research Fellowship (2021) NSF CAREER Award (2020) ONR Young Investigator Award (2020) Grants & Advising: Udell leads projects funded by ONR and NSF, focusing on randomized numerical linear algebra and automated machine learning. She advises PhD students in optimization modeling, deep learning, and causal inference. Her group collaborates on tools like PINNs for scientific computing and OpenReviewer for bias-aware academic reviews. Labs & Teams: She directs the Udell Research Group at Stanford, emphasizing interdisciplinary work in optimization, machine learning, and data science. Collaborations span academia and industry, including work with Aurora Solar on photovoltaic array design and Google Brain on scalable AutoML.

Ali HashemiView profile

Researcher

• Machine Learning
• Statistics
• Optimization
• +3 more

Ali Hashemi is a Researcher at the Berlin Institute for the Foundations of Learning and Data (BIFOLD) and a Postdoctoral Researcher at Technical University of Berlin. He holds a PhD (Dr. rer. nat.) with distinction from TU Berlin (2022) and an MSc in Electrical Engineering from Sharif University of Technology (2013). His research focuses on machine learning, Bayesian methods, and neuroimaging applications, particularly in inverse problems and explainable AI. He has contributed to advancements in structured noise modeling, robust source localization, and 3D neural networks for brain imaging. His work bridges signal processing and AI to improve biomedical data analysis. Education: PhD in Machine Learning, Technical University of Berlin (2022) MSc in Electrical Engineering, Sharif University of Technology (2013) Research Interests: Machine Learning & Optimization Bayesian Inference Neuroimaging (M/EEG, fMRI) Explainable AI & Diffusion Models Signal Processing for Biomedical Applications Key Achievements: Developed the 'Champagne Algorithm' for noise estimation in brain imaging Published widely in top venues like NeurIPS and NeuroImage Recipient of PhD distinction award for outstanding dissertation Labs & Affiliations: Active in TU Berlin's Uncertainty & Inverse Modeling group and Charité’s Brain and Data Science lab. Collaborates on open-source tools for M/EEG analysis (e.g., contributions to MNE-Python).

Pierre HumbertView profile

Researcher

• Conformal Prediction
• Statistical Learning
• Non-Parametric Statistics
• +6 more

Pierre Humbert is a postdoctoral researcher at Sorbonne Université's Laboratoire de Probabilités, Statistique et Modélisation (LPSM), affiliated with the MARS project. He previously held postdoctoral positions at Laboratoire de Mathématiques d'Orsay (LMO) and the INRIA Celeste team. Humbert completed his PhD in 2021 at ENS Paris-Saclay, focusing on multivariate analysis with tensors and graphs in neuroscience under Professors Nicolas Vayatis, Laurent Oudre, and Julien Audiffren. His research interests span conformal prediction, statistical learning, non-parametric methods, robust statistics, signal processing, and applications in neuroscience. He contributes to federated learning frameworks, graph signal processing, and tensor-based methodologies for analyzing complex data structures. Humbert's recent work emphasizes federated conformal prediction for privacy-preserving distributed learning and robust statistical techniques. His publications frequently address graph Laplacian estimation, EEG signal processing, and tensor decomposition applications in biomedical contexts. He collaborates actively with teams like INRIA Celeste and maintains open-source implementations of his algorithms.

• Interpretability
• Machine Learning
• Data Mining
• +4 more

Martino Ciaperoni is a Postdoctoral Researcher and Adjunct Professor in the Department of Computer Science at Aalto University. He is affiliated with the research group of Adj. Prof. Gionis Aris. His primary focus is on advancing interpretable machine learning, algorithm optimization, and data mining techniques. Ciaperoni holds a doctoral degree (Tekn. toht.) in Computer Science from Aalto University (2024) and a Doctoral degree in Engineering and Technology from Università degli Studi di Roma 'La Sapienza' (2019). His research emphasizes interpretability in AI, multi-label classification, and efficient algorithms for speech recognition and Bayesian networks. Recent work includes exploring the Rashomon set of rule-based models, developing low-memory Viterbi decoding algorithms (SIEVE), and solving the Hadamard decomposition problem. His research spans topics like core decomposition in temporal networks and low-rank matrix approximation. Collaborations include international projects in computational linguistics, knowledge discovery, and algorithm design. He has contributed to open-source software, such as the 'Efficient Exploration of the Rashomon Set' codebase (Zenodo). His work balances theoretical rigor with practical applications, aiming to bridge gaps between complex algorithms and real-world problem-solving.

• Numerical Linear Algebra
• Differential-Algebraic Equations
• Control Theory
• +3 more

Prof. Tatjana Stykel is a Professor of Computational Mathematics at the University of Augsburg, affiliated with the Faculty of Mathematics, Natural Sciences, and Materials Engineering. She holds positions in the Institute of Mathematics and leads the Computational Mathematics research group. Her academic journey includes a PhD from Technische Universität Berlin (2002) and a Habilitation (2009), both in numerical mathematics. Prior to her current role, she led a junior research group at TU Berlin’s MATHEON Research Center and held visiting professorships at TU Berlin and Humboldt-Universität zu Berlin. Stykel’s research focuses on Numerical Linear Algebra (e.g., generalized eigenvalue problems, matrix equations), Differential-Algebraic Equations (numerical methods and stability), and Control Theory (descriptor systems, model reduction). She has pioneered energy-adaptive optimization methods and Riemannian geometry techniques for complex systems, including applications in quantum chemistry and electromagnetism. Her awards include the Richard von Mises Prize (2007), Leslie Fox Prize (2003), and multiple fellowships from the Pacific Institute for the Mathematical Sciences and Soros programs. She has led major projects like the EU-funded EU-MORNET (2014–2018) and contributed to BMBF networks in model reduction for coupled systems. Stykel teaches advanced numerical methods, including courses on differential equations, optimization, and model reduction. Her work emphasizes interdisciplinary collaboration, with partnerships in robotics, semiconductor technology, and fluid dynamics. She advises several students, including alumni who have become prominent in applied mathematics and engineering.

Dan AlistarhView profile

Professor

• Machine Learning
• Algorithms
• Distributed Systems
• +3 more

Dan Alistarh is a Professor at the Institute of Science and Technology Austria (ISTA) and ML Research Lead at Neural Magic, Inc. His research focuses on efficient algorithms and systems for machine learning, spanning theoretical foundations to practical implementations. He holds a PhD from EPFL and has held postdoctoral positions at MIT CSAIL and researcher roles at ETH Zurich and Microsoft Research. Education: PhD in Computer Science from EPFL (advisor: Prof. Rachid Guerraoui), Postdoctoral Associate at MIT CSAIL (advisor: Prof. Nir Shavit). Research interests include distributed systems, optimization algorithms, neural network compression (quantization, pruning), and parallel computing. His work bridges theory and practice, with impactful contributions to LLM efficiency, such as GPTQ, Marlin, and SparseGPT. He leads a lab at ISTA with active collaborations and visiting researchers from top institutions. Key scientific awards include the Best Paper Award at DISC 2021 (for work on leader election algorithms) and a Distinguished Paper Award at SPAA 2023. His research is supported by grants from FWF BILAI, ERC, NVIDIA, Google, and Amazon. Advising and grants: Supervises PhD students and postdocs focusing on ML efficiency. Open positions exist for interns, PhD candidates, and postdocs with backgrounds in CS/Math. Recent lab achievements include 4 NeurIPS 2024 acceptances, including an oral presentation for PV-Tuning. Labs/Teams: IST-DASLab, known for open-source contributions (GitHub: IST-DASLab) and innovations in LLM compression. Collaborations span academia and industry, with alumni progressing to roles at OpenAI, Neural Magic, and top universities.

Kemal RoseView profile

Researcher

• Algebraic Geometry
• Optimization
• Polynomial Systems
• +3 more

Kemal Rose is a Postdoctoral researcher at KTH Royal Institute of Technology in Sweden, mentored by professor Sandra di Rocco. He holds a PhD from the Max Planck Institute for Mathematics in the Sciences (Leipzig), advised by Simon Telen and Bernd Sturmfels. His research focuses on algebraic geometry and optimization, with contributions to polynomial systems, tropical geometry, and computational methods in algebraic geometry. Key research themes include certification of polynomial system zeros, p-adic and real cubic surfaces, polyhedral homotopy algorithms, and the algebraic degree of sparse optimization problems. His work bridges theoretical mathematics with practical computational tools, emphasizing interdisciplinary applications in symbolic computation and geometric modeling. Publications span topics such as tropical implicitization, polyhedral-type analysis, and toric geometry. Current research trends reflect a focus on leveraging algebraic methods to solve high-dimensional optimization problems with sparse structures. No scientific awards or grants are explicitly listed in the provided materials. His academic advising history is not detailed here.

Jörn BehrensView profile

Professor

• Numerical Methods in Geosciences
• Climate Modeling
• Tsunami Simulation
• +5 more

Prof. Jörn Behrens is a Professor of Numerical Methods in Geosciences at the University of Hamburg's Department of Mathematics. His work focuses on computational geoscience, including adaptive numerical methods, tsunami modeling, and multiscale simulations. He leads the Numerical Methods in Geosciences research group and is affiliated with the Climate, Climatic Change, and Society (CLICCS) cluster of excellence. Education & Positions: Since 2009: Full Professor at University of Hamburg 2012: Visiting Scientist at Isaac Newton Institute, Cambridge 2007–2009: Privatdozent at University of Bremen Extensive experience at Alfred-Wegener-Institut and TU Munich Research Interests: Behrens develops numerical algorithms for geophysical flows, emphasizing adaptive mesh refinement, discontinuous Galerkin methods, and tsunami hazard modeling. His work bridges computational mathematics and geoscience applications, with a focus on environmental challenges like climate modeling and disaster risk assessment. Recent efforts include cloud computing for instant hazard simulations and interdisciplinary projects like the Global Tsunami Model (GTM). Key Publications: Recent work includes bathymetry reconstruction via PDE-constrained optimization, low-rank preconditioning for fluid flow, and tsunami hazard frameworks within EPOS. These reflect his expertise in computational methods and geophysical problem-solving. Grants & Projects: Lead on EU-funded DT-GEO (Digital Twin for Geophysical Extremes) Co-chair of COST Action AGITHAR (Tsunami Hazard Research) Contributions to CLICCS and EPOS infrastructure development Labs & Teams: Leads the Numerical Methods in Geosciences group, collaborating internationally on tsunami modeling, climate systems, and high-performance computing.

• Artificial Intelligence
• Machine Learning
• Natural Language Processing
• +6 more

Shwai He is a PhD student and Affiliate Assistant Professor at the University of Maryland, advised by Ang Li. Their research focuses on advancing AI systems through innovative approaches in large language models (LLMs), fairness in machine learning, and efficient neural network architectures. Key areas include multi-agent systems, causal modeling for bias mitigation, and optimization techniques for mixture-of-experts models. Research interests span artificial intelligence, machine learning, and natural language processing with emphasis on practical applications like healthcare diagnostics and social pairing systems. Notable work includes developing GNWT-based multi-agent digital twins for social platforms and improving LLM transparency through token analysis. Publications highlight contributions to counterfactual fairness, dynamic-depth transformers, and parameter-efficient methods. Current efforts explore computational efficiency in vision-language models and bio-inspired antibody prediction systems. No scientific awards have been mentioned. Advising and grants: Currently a PhD student under Ang Li's supervision. Research involves collaborations across computer science and bioinformatics domains.

Nick VannieuwenhovenView profile

Assistant Professor

• Tensor decompositions
• Numerical analysis
• Riemannian optimization methods
• +2 more

Nick Vannieuwenhoven is an Assistant Professor at KU Leuven, affiliated with the Department of Computer Science and the NUMA Division. He serves as the Exchange Coordinator for the Master in Mathematical Engineering and is an Associate Editor for The Electronic Journal of Linear Algebra and SIAM Journal on Applied Algebra and Geometry . His research focuses on tensor decompositions, numerical analysis, Riemannian optimization, and applications in data science. He obtained his PhD in 2015 under Professors Karl Meerbergen and Raf Vandebril, funded by the FWO (Research Foundation Flanders). His postdoctoral research (2015–2021) was also supported by FWO fellowships. His research group investigates tensor decompositions, multilinear algebra, and numerical techniques for data science, with a focus on condition number analysis and Riemannian optimization. Collaborators include experts like Carlos Beltrán, Paul Breiding, and Simon Telen. Current students include Jana Jovcheva, Bram Leys, and David Thorsteinsson, working on manifold-valued function approximation, group-invariant networks, and data-based engineering. Key awards include FWO fellowships for his PhD and postdoctoral studies. Grants include support for postdoctoral researchers via MSCA and FWO schemes. Notable projects involve Tucker compression libraries (ATC) and geometric analysis of tensor networks. His work bridges algebraic geometry, numerical analysis, and machine learning, emphasizing stability and computational efficiency.

Kate LarsonView profile

Professor

• Artificial Intelligence
• Multiagent Systems
• Reinforcement Learning
• +3 more

Kate Larson is a Professor at the David R. Cheriton School of Computer Science, University of Waterloo. Her research explores artificial intelligence with emphasis on multiagent systems, reinforcement learning, and AI applications for sustainability and climate challenges. She employs game-theoretic approaches to address coordination and cooperation in complex systems. Education Ph.D. in Computer Science from Carnegie Mellon University (2004), M.Sc. from Washington University in St. Louis (1999), and B.Sc. from Memorial University of Newfoundland (1997). Research Focus Professor Larson's work bridges theoretical AI and practical societal impacts. Key areas include: Designing cooperative AI frameworks for multiagent environments Developing value-alignment mechanisms for ethical AI Applying reinforcement learning to climate change mitigation Creating robust AI systems through game-theoretic guarantees Publication Trends Her recent publications (2024-2025) demonstrate strong focus on AI safety, game-theoretic foundations, and multimodal learning systems. Dominant themes include: algorithmic alignment techniques, multi-agent risk management, and interpretable AI architectures, with increasing emphasis on real-world climate and sustainability applications. Awards & Recognition No major scientific awards mentioned in source materials. Research Infrastructure No specific labs or teams detailed in available information.

Nicolas VAYATISView profile

Professor

• Machine Learning
• Healthcare Technology
• Theoretical Computer Science
• +3 more

Nicolas VAYATIS is a Professor at the Ecole normale supérieure Paris-Saclay, affiliated with the Centre Borelli and the Department of Mathematics (DER de Mathématiques). He holds an ELLIS Fellowship (Paris unit) and specializes in theoretical foundations of machine learning, with applications to healthcare and industrial systems. He teaches in programs such as the Master M2 MVA, the normalien DER mathématiques, and the AI track at ENS Paris-Saclay. His research focuses on statistical learning, domain adaptation, change-point detection, and graph signal processing. He has advised over 20 PhD/Master’s students, including notable names like Antoine de Mathelin and Charles Truong. His work bridges academic theory and practical applications, collaborating on projects like fall detection using smart floor sensors and gait analysis with inertial measurement units. He contributes to open-source tools like the ruptures library for change-point detection and the ADAPT domain adaptation toolbox. Key collaborations include projects with INSEAD, Université Paris-Saclay, and industry partners like Michelin. His research has been published in venues such as NeurIPS, ICML, and AISTATS, with a focus on robust algorithms and real-world impact.

Volker MehrmannView profile

Professor

• Applied and Numerical Linear Algebra
• Analysis and Numerics of Differential-Algebraic Equations
• Numerical Solution of Differential Equations
• +3 more

Volker Mehrmann is a Professor of Mathematics (Numerical Mathematics) at Technische Universität Berlin since 2000. He holds a C4-Professorship and has held leadership roles in major research initiatives such as the DFG Research Center MATHEON and the Einstein Center ECMath. His research focuses on applied and numerical linear algebra, differential-algebraic equations, control theory, and industrial mathematics. Mehrmann earned his Ph.D. from Universität Bielefeld (1982) and habilitation (1988). He has authored influential textbooks, including Lineare Algebra , and serves as Editor-in-Chief of Linear Algebra and its Applications . He is a SIAM Fellow, ERC Advanced Grant recipient, and member of prestigious academies like acatech and Academia Europaea. Education: Diplom (1979), Ph.D. (1982), Habilitation (1988) from Universität Bielefeld. Key Roles: Speaker of DFG Research Center MATHEON (2008–2014) President of GAMM (2011–2016) and EMS (2017–2022) Research Highlights: Model reduction, port-Hamiltonian systems, and energy-based modeling Control of complex systems and industrial applications His recent work emphasizes interdisciplinary projects in energy systems (e.g., BMBF-EiFer, Hybrid Energy Networks) and numerical methods for large-scale systems. Mehrmann’s contributions span theoretical advancements and practical implementations, bridging fundamental mathematics with real-world challenges.