Find Academic Supervisors

Jonathan A. KelnerView profile

Professor

• Algorithms
• Spectral Graph Theory
• Combinatorial Optimization
• +7 more

Jonathan A. Kelner is a Professor of Applied Mathematics in the Department of Mathematics at the Massachusetts Institute of Technology (MIT) and a member of the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL). His research focuses on applying techniques from pure mathematics to solve fundamental problems in algorithms and complexity theory, with the goal of developing practical algorithms for real-world questions. Dr. Kelner received his undergraduate degree from Harvard University and his Ph.D. in Computer Science from MIT in 2006. Before joining the MIT faculty, he spent a year as a member of the Institute for Advanced Study. His educational background has provided a strong foundation for his interdisciplinary research spanning mathematics and computer science. His research interests include combinatorial optimization, mathematical programming, spectral graph theory, distributed computing, machine learning, computational geometry and topology, computational biology, signal processing, and random matrix theory. Kelner's work demonstrates how deep theoretical insights can lead to practical algorithmic improvements, particularly in graph algorithms and optimization problems. His approach often involves connecting seemingly disparate areas of mathematics to create novel algorithmic techniques. Analysis of his recent publications reveals a strong focus on spectral graph theory, optimization algorithms, and the Sum-of-Squares method. His work frequently addresses fundamental questions in theoretical computer science with practical implications for algorithm design. A notable trend in his research is the development of nearly-linear-time algorithms for various graph problems, which represents significant improvements over previous approaches. NSF CAREER Award Alfred P. Sloan Research Fellowship NEC Award for Research in Computers and Communication Sprowls Doctoral Dissertation Award Best Student Paper Award at STOC 2004 Best Paper Award at STOC 2011 Kokusai Denshin Denwa Junior Faculty Chair 2008 Harold E. Edgerton Faculty Achievement Award 2011 School of Science Award for Excellence in Undergraduate Education 2012 Professor Kelner has been actively involved in mentoring students and has received recognition for his teaching excellence, including the School of Science Award for Excellence in Undergraduate Education in 2012. His research has been supported by prestigious grants including the NSF CAREER Award. He has collaborated extensively with researchers across multiple institutions, often working with other leading figures in theoretical computer science to produce groundbreaking results in algorithm design. At MIT, Kelner is part of both the Mathematics Department and CSAIL, positioning him at the intersection of theoretical mathematics and practical computer science. This dual affiliation reflects the interdisciplinary nature of his work, which bridges pure mathematical theory with concrete algorithmic applications.

Ivan DokmanicView profile

Assistant Professor

• Machine Learning
• Signal Processing
• Inverse Problems
• +11 more

Ivan Dokmanic is an Assistant Professor at the Coordinated Science Laboratory (CSL) within the University of Illinois . His research bridges signal processing , machine learning , and applied inverse problems , with a focus on acoustics, biomedical imaging, and distance geometry. Current Role : Assistant Professor, CSL Email : dokmanic@illinois.edu Research Interests : Dokmanic explores machine learning applications in inverse problems , particularly distance geometry for molecular imaging and acoustics . His work includes unlabeled sensing , where distances between points are known but their arrangement is not. This has implications for powder diffraction , indoor localization , and echo modeling . Article Trends : His recent publications emphasize distance geometry in machine learning , acoustic signal processing , and inverse problem theory . Key areas include molecular imaging , audio encryption , and sensor positioning . Collaborative work spans medical imaging , cyberphysical systems , and geometric invariants . 2016 Google Faculty Award NSF Grant (1 year, $157,079) Students and Grants : Dokmanic mentors PhD students like Puoya, Shuai, and Anadi. His research is funded by the National Science Foundation , Google , VISA , and nVidia .

Jose Israel RodriguezView profile

Associate Professor

• Applied Algebraic Geometry
• Algebraic Statistics
• Nonlinear Algebra
• +7 more

Jose Israel Rodriguez is an Associate Professor in the Department of Mathematics at the University of Wisconsin-Madison. His research bridges applied algebraic geometry and algebraic statistics, focusing on nonlinear algebra, maximum likelihood estimation, monodromy, and polynomial systems in engineering and science applications. Primary Affiliation: Department of Mathematics , UW-Madison Additional Affiliations: Department of Electrical & Computer Engineering , Institute for Foundations of Data Science Research Interests : Applied algebraic geometry for nonlinear eigenvalue problems and kinematics Algebraic statistics in nearest point problems and likelihood geometry Numerical methods for monodromy, Galois groups, and polynomial optimization Teaching and Mentorship : Co-organized the Collaborative Undergraduate Research Laboratory (CURL) for Spring 2020 Advises PhD students Julia Lindberg and Zinan Wang , with Bernd Sturmfels as his own PhD advisor Developed software tools like Multiregeneration and Decomposable Sparse Polynomial Systems Academic Contributions : Authored over 20 peer-reviewed publications in journals like SIAM Journal on Applied Algebra and Geometry, Foundations of Computational Mathematics, and Journal of Symbolic Computation Organized international conferences including Monodromy and Galois Groups in Enumerative Geometry and SIAM AG19 Active member of the SIAM community and developer of the Matroids Day seminar

• Generative AI
• Information Theory
• Machine Learning
• +4 more

Alexandros G. Dimakis is a Professor at the University of California, Berkeley's Department of Electrical Engineering and Computer Sciences (EECS), College of Engineering. He is also Co-Director of the National AI Institute for Foundations of Machine Learning and Co-Founder of BespokeLabs.ai. PhD (2008) and Diploma (2003) in Electrical Engineering His research focuses on Generative AI , Information Theory , and Machine Learning . Recent work includes advancements in diffusion models, compressed sensing, and causal inference. His publications (150+) emphasize inverse problems, neural network verification, and generative model optimization. Recent publications highlight trends in Diffusion Models for inverse problems, Language Model Scaling , and 3D-Aware Generative Systems . Collaborative projects span biomedical applications, large-scale dataset curation (Datacomp-LM), and parameter-efficient model fine-tuning. Scientific Awards : IEEE Fellow (2022) James Massey Award (2018) NSF CAREER Award (2011) Google Research Faculty Award Best Paper awards at UAI workshops Eli Jury Dissertation Award (UC Berkeley) He advises PhD students in generative modeling, compressed sensing, and information theory. His research group collaborates with institutions like MIT, NYU, and IBM Research. Former students hold positions at Google, Amazon, and academic institutions like Purdue University.

Stephen A. VavasisView profile

Professor

• Continuous Optimization
• Data Science
• First-Order Methods
• +4 more

Stephen A. Vavasis is a Professor in the Department of Combinatorics and Optimization at the University of Waterloo, part of the Faculty of Mathematics. He holds a PhD in Computer Science from Stanford University (1989) and has held academic positions at Cornell University (1989–2006) before joining Waterloo. His research focuses on continuous optimization, data science, first-order methods, scientific computing, and computational mechanics. Current teaching includes courses on convex optimization and portfolio optimization methods. He has served as Associate Dean of Computing (2017–2020) and Interim Director of Data Science graduate programs. His work is supported by NSERC grants. Notable awards include the Hertz Fellowship, Churchill Scholarship, and Guggenheim Fellowship. Vavasis's research emphasizes applications of optimization to clustering, machine learning, and fracture mechanics. His publications span convex optimization frameworks, algorithmic analysis of gradient methods, and numerical methods in mechanics. Recent work explores unifying analyses of first-order optimization algorithms and robust optimization techniques for high-dimensional data problems.

Jee ChoiView profile

Assistant Professor

• High Performance Computing
• Parallel and Scalable Algorithm Design
• Performance Analysis Optimization
• +2 more

Jee Choi is an Assistant Professor in the Department of Computer and Information Science within the College of Arts and Sciences at the University of Oregon. His research focuses on developing high-performance algorithms for big data analytics, particularly in tensor decomposition and parallel computing systems. Education: PhD in Electrical and Computer Engineering, Georgia Institute of Technology, 2015 MS in Electrical and Computer Engineering, Georgia Institute of Technology, 2004 BS in Electrical and Computer Engineering, Georgia Institute of Technology, 2000 Research Interests: Dr. Choi specializes in High Performance Computing with expertise in parallel algorithm design, performance modeling, and energy efficiency. His work targets Tensor Decomposition & Data Mining for big data, developing scalable solutions for sparse and dense tensor computations. He advocates leveraging HPC systems to transform massive datasets into societal benefits through efficient data mining techniques. Publication Trends: Choi's recent publications (2021-2025) demonstrate consistent innovation in tensor decomposition methodologies, with increasing focus on adaptive storage (Alto), streaming data factorization, and energy-aware autotuning. His work spans GPU clusters, distributed systems, and emerging architectures, maintaining strong connections to real-world big data challenges while advancing theoretical algorithm design. Professional Background: After completing his PhD under Richard Vuduc at Georgia Tech (notable for SpMV GPU autotuning and Energy Roofline Model), Choi conducted DoD-funded research on tensor decomposition at IBM Watson Research Center (2015-2018) before joining the University of Oregon faculty.

• Numerical Analysis
• Partial Differential Equations
• Wave Propagation
• +7 more

Olof Runborg is a Professor in Numerical Analysis at the Royal Institute of Technology (KTH) in Stockholm, Sweden. He works in the Division of Numerical Analysis, Optimization and Systems Theory within the Department of Mathematics at KTH. His research focuses on developing and analyzing numerical methods for partial differential equations, particularly for wave propagation problems. His educational background includes: MSc in Electrical Engineering from KTH (1992) BSc in Economics & Business Administration from SSE (1994) PhD in Numerical Analysis/Applied Mathematics from KTH (1998) Postdoc at Paris VI University (1999) Postdoc at Princeton University's Program in Applied and Computational Mathematics (2000-2001) Became Docent at NADA in 2003 Appointed Professor at KTH in 2010 Professor Runborg's research centers on the numerical treatment of partial differential equations, with special emphasis on wave propagation problems. His work spans multiple areas including high-frequency waves, multiscale phenomena, numerical homogenization, uncertainty quantification, multiresolution analysis, Gaussian beams, and mesh generation. A recurring theme in his research is developing methods to solve computationally expensive problems more efficiently while maintaining accuracy. His approach often involves coupling different numerical methods or reformulating equations to create more efficient computational approaches. His research has applications across physics and engineering domains where wave phenomena are important. An analysis of his recent publications (2015-2025) shows continued focus on high-frequency wave propagation with expanding applications to areas like the Landau-Lifshitz equation for magnetic materials and elastic wave propagation. His work bridges theoretical numerical analysis with practical computational techniques, often developing novel methods like the WaveHoltz iteration for solving the Helmholtz equation. The publications demonstrate increasing attention to uncertainty quantification and multiscale methods while maintaining strong theoretical foundations in error analysis. Professor Runborg teaches several courses at KTH including Numerical Methods (basic course), Applied Numerical Methods, Numerical Algorithms for Data-Intensive Science, and Selected Topics in Numerical Analysis II. He serves as examiner for degree projects in Scientific Computing. His teaching reflects his research expertise in numerical methods and computational mathematics, providing students with both theoretical foundations and practical implementation skills. Beyond his core research, Professor Runborg has engaged in interesting side projects, such as his analysis of 'Numbers on the Web' where he investigated the frequency of numbers 11-1000 in web content, revealing patterns related to dates, time, computer systems, and cultural phenomena. This demonstrates his broader interest in data analysis and computational approaches to understanding patterns in information.

Euiwoong LeeView profile

Assistant Professor

• Approximation Algorithms
• Hardness of Approximation
• Discrete Optimization
• +3 more

Euiwoong Lee is an Assistant Professor in the Computer Science and Engineering Division at the University of Michigan. He holds a PhD from Carnegie Mellon University, advised by Venkatesan Guruswami, and has held postdoctoral positions at New York University and the Simons Institute for the Theory of Computing. His research focuses on approximation algorithms, hardness of approximation, and parameterized complexity. **Education:** PhD in Computer Science, Carnegie Mellon University (2017), advised by Venkatesan Guruswami Postdoctoral Fellowships: NYU (2017–2020), Simons Institute (2017–2020) **Research Interests:** Approximation Algorithms & Hardness of Approximation Convex Hierarchies (e.g., Sum-of-Squares) Clustering Algorithms (e.g., Correlation Clustering) Parameterized Complexity Facility Location & Metric Optimization **Awards:** Edmund M. Clarke Doctoral Dissertation Award (2017) Simons Award for Graduate Students in Theoretical Computer Science **Advising & Grants:** PhD Students: Anthony Della Pella, Aditya Anand, Amatya Sharma, Ian DeHaan Co-organizes the Michigan Theory Seminar **Labs/Teams:** Collaborates with researchers in approximation algorithms, optimization, and theoretical computer science at the University of Michigan and beyond.

Chris De SaView profile

Associate Professor

• Machine Learning
• Distributed Systems
• Optimization Algorithms
• +5 more

Chris De Sa is an Associate Professor in the Department of Computer Science at Cornell University, affiliated with the Cornell Machine Learning Group and leading the Relax ML Lab. His research focuses on algorithmic, software, and hardware techniques for high-performance machine learning, particularly relaxed-consistency stochastic algorithms like asynchronous and low-precision stochastic gradient descent (SGD). He earned his Ph.D. from Stanford University under advisors Kunle Olukotun and Chris Ré. His work emphasizes constructing efficient, parallel, and distributed machine learning frameworks for deep learning and data analytics. Education: Ph.D. in Computer Science, Stanford University (2017) Research Interests: Algorithmic techniques for scalable ML, quantization, distributed optimization, hyperbolic geometry in ML, and reliable measurement of ML systems. His group develops frameworks for efficient inference/training and explores the intersection of ML with domains like agriculture and plant science through courses like PLSCI 7202. Recent Highlights: DARPA YFA Grant (2024), NSF CAREER Award, Google Research Scholar Award, and multiple best paper recognitions. Key contributions include QuIP quantization methods, Coneheads attention mechanisms, and theoretical advances in decentralized training. Awards: NSF CAREER Award DARPA YFA Grant (2024) Google Research Scholar Award Mr. & Mrs. Richard F. Tucker Teaching Award Grants & Advising: Advises 8 Ph.D. students (including Ruqi Zhang, Yucheng Lu, A. Feder Cooper) and holds leadership roles in MLSys conferences. Active in grant-funded research (e.g., NSF Robust Intelligence). Labs/Teams: Leads the Relax ML Lab and participates in Cornell’s Institute for Digital Agriculture (CIDA).

• High-Performance Computing
• Reconfigurable Computing
• FPGA Acceleration
• +7 more

Carlos Alvarez Martinez is a faculty member at the Universitat Politècnica de Catalunya (UPC), affiliated with the Department of Computer Architecture at the Barcelona School of Informatics (FIB). He is a key member of the Programming Models (PM) research group and collaborates closely with the Barcelona Supercomputing Center (BSC). His research focuses on high-performance computing, FPGA acceleration, task-based programming models like OmpSs, and hardware-software co-design for heterogeneous systems. His research interests center on advancing parallel computing through innovative programming models and hardware acceleration. He investigates efficient task scheduling, resource management in multicore and FPGA-based systems, and runtime support for dataflow models. His work enables high-performance execution of complex applications in domains such as scientific computing and cyber-physical systems. He actively contributes to European initiatives like TEXTAROSSA and AXIOM, aiming to develop next-generation exascale supercomputing technologies. The trend in his recent publications shows a strong focus on leveraging FPGAs for HPC, optimizing SpMV operations, improving task scheduling with hardware support, and developing frameworks for multi-FPGA clusters. His work consistently bridges theoretical models with practical implementations, emphasizing performance, scalability, and energy efficiency in heterogeneous computing environments. Scientific Awards: Premi UPC al Compromís Social 2019 Premi Disseny per al Reciclatge 2013 Alvarez Martinez has advised or collaborated with several doctoral students, including Jaume Bosch, Xubin Tan, and Fahimeh Yazdanpanah. He has been involved in numerous competitive R&D projects, often related to high-performance computing and parallel programming models. His work includes significant contributions to educational innovation, particularly in active learning methodologies and formative assessment using interactive systems. He leads and participates in research labs and teams focused on programming models and computer architecture, notably the PM group at UPC/BSC. These teams develop runtime systems, compilers, and hardware accelerators to push the boundaries of parallel computing efficiency and programmability.

• VLSI Design
• Computer-Aided Design
• Spectral Graph Theory
• +4 more

Zhuo Feng is Professor of Electrical and Computer Engineering at Stevens Institute of Technology, directing the HUDSON Lab and holding a Ph.D. from Texas A&M University. His research develops spectral graph methods for VLSI design, including circuit simulation, power grid verification, and machine learning applications. Funded by NSF CAREER and multiple grants, his work has produced award-winning algorithms like GRASS for graph sparsification. Recent publications focus on spectral methods for circuit stability analysis, physics-informed neural networks, and explainable AI frameworks. He teaches graduate courses in VLSI design and GPU programming while co-founding LeapLinear Solutions. NSF CAREER Award (2014) ACM/IEEE DAC Best Paper Award (2013) Multiple Best Paper Nominations (ICCAD 2008, 2006)

Elette BoyleView profile

Associate Professor

• Cryptography
• Computer Security
• Coding Theory
• +7 more

Elette Boyle is an Associate Professor at Reichman University (IDC Herzliya) and a Senior Scientist at NTT Research . She holds a Ph.D. in Mathematics from MIT (advised by Shafi Goldwasser and Yael Tauman Kalai) and an undergraduate degree from Caltech . Education Ph.D. in Mathematics, MIT B.S. in Mathematics, Caltech Her research focuses on cryptographic solutions for secure data processing , particularly in secure multi-party computation , function/homomorphic secret sharing , and distributed point functions . Recent work explores topology-hiding communication , memory checking complexity , and sublinear-communication MPC . Key trends in her publications include: Advancements in Function Secret Sharing for branching programs and sparse vectors. Efficient Secure Multi-Party Computation protocols with preprocessing. Information-theoretic and computational Topology-Hiding Broadcast schemes. Optimized Oblivious Transfer with constant computational overhead. Scientific Awards European Research Council (ERC) Award Israeli Science Foundation (ISF) Grant United States Air Force Office of Scientific Research (AFOSR) Grant Google Research Scholar Award International Association for Cryptologic Research (IACR) Recognition As Director of the Foundations & Applications of Cryptography (FACT) Research Center , she leads collaborative work with institutions like Technion Israel , Cornell University , and NTT Research . Her students include Pierre Meyer (Ph.D.) , Matan Hamilis (Ph.D.) , and D'or Banon (MSc.) .

Clyde KruskalView profile

Associate Professor

• Parallel Computing
• Algorithms
• Computer Architecture
• +3 more

Clyde Kruskal is an Associate Professor in the Department of Computer Science at the University of Maryland, College Park. His research focuses on parallel architectures, models, and algorithms. He earned a Ph.D. from New York University in 1981 and a bachelor’s degree from Brandeis University in 1976. His work includes foundational contributions to parallel computing, such as the read–modify–write concept in distributed systems. Kruskal’s research spans topics like interconnection networks, synchronization mechanisms, and algorithm design for parallel systems. Education: Bachelor’s Degree: Brandeis University, 1976 Master’s Degree: New York University (Courant Institute), 1978 Ph.D.: New York University (Courant Institute), 1981 Research Interests: Parallel computing architectures, parallel algorithms design, multiprocessor synchronization, interconnection networks, and computational geometry problems like graph coloring and visibility analysis. His work emphasizes theoretical foundations and practical implementations in parallel systems. Notable Contributions: Kruskal co-authored the book Problems With A Point: Exploring Math And Computer Science (2019), and his research includes foundational papers on parallel prefix operations, sparse matrix algorithms, and synchronization protocols. His publications span over three decades, reflecting sustained contributions to parallel computing theory and practice. Advising & Outreach: He has mentored students through programs like the Summer Combinatorial Algorithms REU at UMD, fostering undergraduate research in algorithm design and parallel computing.

Pengtao XieView profile

Associate Professor

• Machine Learning
• Natural Language Processing
• Machine Learning for Healthcare
• +4 more

Pengtao Xie is an Associate Professor (tenured) in the Department of Electrical and Computer Engineering at UC San Diego, with cross-appointments in the Division of Biomedical Informatics and affiliations across multiple schools and institutes including the Halıcıoğlu Data Science Institute, School of Biological Sciences, and Skaggs School of Pharmacy. His research bridges human-inspired machine learning and healthcare applications. Education: PhD in Machine Learning, Carnegie Mellon University (2018) MS from Tsinghua University BS from Sichuan University Research Interests: His work focuses on machine learning inspired by human learning strategies , including learning by testing, interleaving, self-explanation, and teaching. These techniques are applied to large language models , foundation models , healthcare , and biomedicine . Recent efforts emphasize generative AI for medical image segmentation and protein function prediction. Scientific Awards: NIH MIRA Award (2025) NSF Career Award (2024) Best Graduate Teacher Award, ECE UCSD (2023) ICLR Notable-Top-5% Paper (2023) Global Top-100 Chinese Young Scholars in AI (2022) Tencent Faculty Award (2021) AMIA Doctoral Dissertation Award Finalist (2020) Siebel Scholarship (2014) Teaching & Mentorship: He has developed and taught courses such as Deep Generative Models , Probabilistic Graphical Models , and Linear Algebra and Applications . He actively mentors PhD, master's, and undergraduate students, with alumni now at CMU, Stanford, Mila, and industry roles. Labs & Teams: He leads a research group within the Center for Machine-Intelligence, Computing and Security and collaborates with the Institute for Genomic Medicine and Institute of Engineering in Medicine at UC San Diego.

Mathias DrtonView profile

Professor

• Graphical models
• Algebraic statistics
• Causal inference
• +3 more

Prof. Mathias Drton holds the Chair of Mathematical Statistics at the Technical University of Munich (TUM), within the Department of Mathematics and School of Computation, Information and Technology. His research focuses on graphical models, algebraic statistics, causal inference, and multivariate data analysis. He has authored numerous publications in top-tier journals and conferences, including work on conditional independence, sparse factor analysis, and causal discovery in linear models. Drton has supervised a large number of theses, mentoring students in areas like high-dimensional statistics, graphical models, and causal inference. He is actively involved in teaching advanced courses such as 'Graphical Models in Statistics' and 'Fundamentals of Mathematical Statistics.' His academic contributions span theoretical developments in statistical methodology and computational tools, including R packages like SEMID and symRC . Drton collaborates internationally, contributing to projects like the TUM-ICL Mathematical Sciences Hub and Exzellenzcluster MCQST. His work bridges algebraic methods with statistical challenges, addressing identifiability in latent variable models and robust graphical modeling under non-Gaussian assumptions. Recent research emphasizes causal structure learning under partial homoscedasticity, distribution-free independence tests, and multi-domain causal representation learning. Drton’s lab actively explores applications in genomics, epidemiology, and machine learning, leveraging both theoretical rigor and practical computational methods.