Anoosheh Heidarzadeh is an Assistant Professor in the Department of Electrical and Computer Engineering at Santa Clara University's School of Engineering. He holds a Ph.D. in Electrical and Computer Engineering from Carleton University (2012) and previously served as a Visiting Assistant Professor at Texas A&M University (2018-2022) and Associate Research Scientist at the same institution (2015-2017). His postdoctoral research was conducted at the California Institute of Technology (2013-2014). His research focuses on: Information and coding theory : Fundamental limits of data transmission and storage systems Private and secure computing : Protocols for confidential data processing in networked environments Fault-tolerant distributed systems : Resilient computation frameworks for large-scale applications Distributed machine learning : Scalable algorithms for collaborative learning architectures Recent publications (2021-2022) demonstrate strong emphasis on privacy-preserving computation (covering 73% of articles) and distributed coding techniques (67% of articles), with innovations in private information retrieval, matrix operations, and group testing methodologies. Theoretical contributions dominate (87%), while 13% address applied challenges like COVID-19 screening.
Michael O'Boyle is a Professor at the University of Edinburgh's School of Informatics, where he serves as Director of the ARM Research Centre of Excellence and the EPSRC Centre for Doctoral Training in Pervasive Parallelism. Holding an EPSRC Established Career Research Fellowship, he leads pioneering work in compiler technology for heterogeneous architectures, bridging theoretical advances with practical high-performance computing applications. Professor O'Boyle's research spans multiple cutting-edge areas including heterogeneous code discovery and optimization, neural machine translation for program synthesis, deep neural network system stack optimization, software-defined hardware, and compiler/architecture co-design. His approach integrates constraint analysis, program synthesis, and machine learning to address complex challenges in high-performance computing across diverse hardware platforms. His recent publications reveal a strong trend toward integrating machine learning with traditional compiler techniques, particularly in neural program synthesis, tensor optimization, and architecture-aware compilation. This work represents a paradigm shift in compiler design, moving from rule-based systems to learning-based approaches that can automatically adapt to diverse hardware targets. IEEE/ACM CGO 2025 Distinguished Paper Award for 'Tensorize: Fast Synthesis of Tensor Programs from Legacy Code' IEEE/ACM CGO 2024 Test of Time Award ACM GPCE 2023 Best Paper Award for 'C2TACO: Lifting Tensor Code to TACOM' ACM ASPLOS 2021 Distinguished Paper Award IEEE HPCA 2021 Best Paper Award for 'Prodigy: Improving the Memory Latency of Data-Indirect Irregular Workloads' Professor O'Boyle has successfully mentored numerous PhD students who have secured prominent positions in academia (including at Cambridge, Edinburgh, Leeds, and McGill) and industry (including Meta, NVIDIA, Qualcomm, Huawei, and Microsoft). His research is supported by significant funding from EPSRC, ARM, and European projects including Bonseyes and Transmuter, demonstrating strong international recognition and industry impact. He leads the influential Compiler and Architecture Design (CArD) Group at the University of Edinburgh and is a founder of the HiPEAC Network of Excellence, which has grown into a major European initiative connecting researchers and practitioners in high-performance and embedded computing.
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.
Dr. Baijian "Justin" Yang serves as the Associate Dean for Research at Purdue Polytechnic Institute and is a Professor in the Department of Computer and Information Technology at Purdue University. He earned his Ph.D. in Computer Science from Michigan State University, with Master's and Bachelor's degrees in Automation (EECS) from Tsinghua University. Dr. Yang has established himself as a leader in multiple interdisciplinary research domains. Dr. Yang's educational background includes: PhD in Computer Science, Michigan State University (2002) MS in Automation (EECS), Tsinghua University (1998) BS in Automation (EECS), Tsinghua University (1995) His research interests span multiple cutting-edge domains with practical applications: Cybersecurity : Developing novel approaches for threat intelligence, security education, and network defense Big Data : Creating innovative algorithms for dimension reduction, regression with categorical variables, and tensor decomposition Applied Machine Learning : Implementing AI solutions in healthcare, manufacturing, and forestry applications Digital Forestry : Using UAV imagery and remote sensing for forest management and tree species classification Dr. Yang's publication record demonstrates significant impact across multiple disciplines, with recent work focusing on spatial transcriptomics analysis (SiGra), delirium detection using limited-lead EEG, and visual localization technologies. His research bridges theoretical advances with practical applications in healthcare, manufacturing quality control, and environmental monitoring. The interdisciplinary nature of his work is evident in collaborations spanning computer science, healthcare, forestry, and manufacturing domains. His scientific achievements have been recognized with numerous awards: 2023 HRSA Building Bridges to Better Health Competition Winner (Phase 1) and 2nd place ($100,000 prize) in Phase 3 2023 Outstanding Faculty Award in Engagement, Department of Computer and Information Technology, Purdue University 2021 Leadership in Manufacturing Award, Manufacturing Times Digital (MxD) 2021 Good to Great Award, Purdue Polytechnic 2020 Outstanding Faculty Award in Discovery, Department of Computer and Information Technology 2019 University Faculty Scholars, Purdue University As an educator and mentor, Dr. Yang has advised numerous graduate students through their PhD and Master's research. His leadership extends to significant service roles including serving as Faculty Champion for the Holistic Safety and Security research impact area at Purdue Polytechnic from 2018 to 2021, board membership with ATMAE (2014-2016), and participation in the IEEE Cybersecurity Initiative Steering Committee (2015-2017). He holds valuable industry certifications including CISSP, MCSE, and Six Sigma Black Belt, demonstrating his commitment to bridging academic research with industry practice. Dr. Yang leads multiple research projects including "Digital Forestry" for developing tools to quantify forest function, "CHEESE" (Cyber Human Ecosystem of Engaged Security Education), and "CICI" (Supporting Controlled Unclassified Information with a Campus Awareness and Risk Management Framework). His work on "Applied Machine Learning" focuses on solving real-world problems, while his "Dimension Reduction and Memory Amnestic Big Data Regression" project innovates computational algorithms for large-scale data analysis.
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.
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.
Venkatesan Guruswami is a Chancellor's Professor in the Department of EECS and a Senior Scientist at the Simons Institute for the Theory of Computing at UC Berkeley . He also holds a Professor position in the Department of Mathematics . His academic journey began with a B.Tech in Computer Science from the Indian Institute of Technology, Madras (1997) , followed by a Ph.D. in Computer Science from the Massachusetts Institute of Technology (2001) . After a Miller Research Fellowship at UC Berkeley (2001–02), he held faculty roles at the University of Washington and Carnegie Mellon University before returning to UC Berkeley in January 2022. Education : B.Tech, IIT Madras (1997) Ph.D., MIT (2001) Professional Affiliations : Chancellor's Professor, UC Berkeley (EECS) Senior Scientist & Interim Director, Simons Institute Professor, UC Berkeley (Mathematics) Guruswami's research spans multiple domains within Theoretical Computer Science , focusing on Error-Correcting Codes , Approximation Algorithms , Randomness in Computing , Probabilistically Checkable Proofs , and Computational Complexity . His groundbreaking work in List Decoding has enabled codes with minimal redundancy for correcting worst-case errors, while recent advancements include Polar Codes , Deletion-Correcting Codes , and Constraint Satisfaction Problems . He has also contributed to Quantum Coding Theory , Locally Recoverable Codes , and Approximation Hardness in various computational contexts. His publications reflect a deep engagement with interdisciplinary topics. Key trends include: Quantum Information Theory : Quantum LDPC codes, transversal gates, and quantum storage. Algebraic Coding : Reed-Solomon codes, AG codes, and polynomial-based constructions. Computational Complexity : Hardness of approximation, CSPs, and parameterized intractability. Data Transmission : Polar codes, deletion channels, and feedback mechanisms. Algorithmic Techniques : Spectral methods, semirandom models, and Lasserre hierarchy applications. Guruswami has received numerous accolades, including the Simons Investigator Award , Presburger Award , Packard Fellowship , Sloan Research Fellowship , ACM Doctoral Dissertation Award , and the IEEE Information Theory Society Paper Award . He is an ACM Fellow (2017) and IEEE Fellow (2019) , with recent honors like the Guggenheim Fellowship (2023) and AMS Fellow (2023) . As an advisor, he has mentored over 25 PhD and postdoctoral researchers , including Atri Rudra , Prasad Raghavendra , and Peter Manohar , whose work has won awards like the Edmund M. Clarke Doctoral Dissertation Award and CRA Outstanding Undergraduate Researcher Award . His research is supported by grants from the National Science Foundation , Packard Foundation , and Sloan Foundation . He also serves as Editor-in-Chief of the Journal of the ACM and holds leadership roles in IEEE and arXiv moderation. Guruswami is actively involved in Simons Institute programs and co-organized workshops on Coded Computation and Information Theory . His work bridges theoretical advancements with practical applications in Cloud Storage , Quantum Computing , and Group Testing , including pandemic-era contributions like AC-DC: Amplification Curve Diagnostics for SARS-CoV-2 .
Garnet K. Chan is the Bren Professor of Chemistry and Director of the Rudolph A. Marcus Center for Theoretical Chemistry at the California Institute of Technology. He received his B.S. from the University of Cambridge in 1996 and his M.A. and Ph.D. from the University of Cambridge in 2000. Dr. Chan's research lies at the interface of theoretical chemistry, condensed matter physics, and quantum information theory, focusing on quantum many-particle phenomena and the numerical methods to simulate them. His group has developed numerous methodologies including density matrix renormalization and tensor network algorithms, canonical transformation-based down-foldings, local quantum chemistry methods, quantum embeddings, and new quantum Monte Carlo algorithms. His work addresses problems that appear naively exponentially hard but where understanding of physics, particularly entanglement structure, allows for calculations of polynomial cost. Analysis of his recent publications reveals a strong focus on quantum simulation techniques, particularly tensor network methods applied to strongly correlated systems. His research spans fundamental theoretical developments to practical applications in quantum computing, molecular simulation, and materials science, with increasing integration of machine learning techniques and GPU acceleration in computational chemistry frameworks. Dr. Chan leads an active research group at Caltech dedicated to simulating chemical and physical systems at the level of many-particle quantum mechanics. His group has welcomed numerous researchers including Kasra Hejazi, Zuxin Jin, Zhihao Cui, Ke Liao, Henrik Larsson, and Wenyuan Liu. He teaches courses in Physical Chemistry (Ch 21 abc) and Advanced Quantum Chemistry (Ch 225), contributing significantly to theoretical chemistry education at Caltech.
Amir Shaikhha is an Associate Professor (Reader) in the School of Informatics at the University of Edinburgh. He was previously an Assistant Professor (Lecturer) at the same institution from 2020 to 2024 and a Departmental Lecturer at the University of Oxford until August 2020. He is also a Junior Research Fellow at University College, Oxford. His academic journey began with a Ph.D. from EPFL in 2018, where he was awarded the Google Ph.D. Fellowship in structured data analysis and a Ph.D. thesis distinction. His research centers on the design and implementation of data-analytics systems, drawing upon techniques from databases, programming languages, compilers, and machine learning. He develops high-performance systems such as SDQL.py, StructTensor, and VecHT, focusing on the compilation of data science workloads and optimization of tensor operations. His work bridges the gap between high-level abstractions and efficient execution, particularly in sparse and probabilistic computing domains. The recent publications highlight a strong trend in compiler-driven optimizations for data-intensive applications, including automatic differentiation, loop fusion, probabilistic programming, and domain-specific language (DSL) restaging. His research integrates machine learning for systems decisions and emphasizes reproducibility and performance. He has published consistently in top venues like PLDI, OOPSLA, SIGMOD, and CGO, reflecting sustained impact in programming languages and database systems. Dahl-Nygaard Junior Prize, 2025 Google Research Scholar Award, 2025 Most Influential Paper Award, GPCE 2024 Best Paper Award, GPCE 2017 Most Reproducible Paper Award, SIGMOD 2017 Google Ph.D. Fellowship, 2017 Amir Shaikhha has advised PhD students including Hesam Shahrokhi and has been nominated for Best Supervisor of the Year at the University of Edinburgh. He leads research projects that have received recognition and support through awards and grants, including the Google Research Scholar Award. He actively serves the community through program committees (e.g., GPCE, DBPL, DRAGSTERS), editorial roles, and peer review for premier journals. His leadership in organizing workshops and conferences underscores his role as a central figure in the programming languages and databases research communities. He leads a research group focused on compiler and database systems, with recent open-source releases such as StructTensor and VecHT. His team collaborates with researchers from institutions like MIT, EPFL, and TU Berlin, and he co-chairs workshops like Sparse@PLDI and DRAGSTERS. His lab emphasizes innovation in how data-intensive programs are compiled and executed efficiently across modern hardware.
David Hong is an Assistant Professor in the Department of Electrical and Computer Engineering at the University of Delaware. He holds a PhD from the University of Michigan, where he was an NSF Graduate Research Fellow, and previously served as an NSF Postdoctoral Research Fellow at the University of Pennsylvania. His research focuses on developing robust methods for analyzing heterogeneous and high-dimensional data, particularly through low-rank matrix and tensor techniques. Applications span medical imaging, radar systems, genomics, and astronomy. He emphasizes theoretical guarantees and practical algorithms for signal extraction and inverse problems. Education: PhD in Electrical Engineering and Computer Science (University of Michigan), NSF Postdoctoral Research Fellowship (University of Pennsylvania). Research Interests: Low-rank matrix/tensor methods, heterogeneous data analysis, unsupervised learning, and applications in healthcare, imaging, and sensor systems. His work addresses noise robustness, scalable algorithms, and real-world deployment challenges. Scientific Awards: Recipient of the NSF Postdoctoral Research Fellowship (2020) and NSF Graduate Research Fellowship (2015). Advising & Grants: Advisor to graduate students in machine learning and signal processing (no named advisees listed). Active NSF grant recipient for foundational and applied research in data science. Labs/Teams: Engaged in interdisciplinary collaborations through the University of Delaware's Center for Computational Research and Data Science initiatives.
David Steurer is an Associate Professor in the Department of Computer Science at ETH Zurich, where he leads the Institute for Theoretical Computer Science. His research bridges theoretical computer science, mathematics, and practical applications in machine learning and statistics. Steurer has made significant contributions to the understanding of sum-of-squares methods, semidefinite programming, and high-dimensional estimation problems. Steurer earned his B.Sc. & M.Sc. in Computer Science from Saarland University in 2006, followed by a Ph.D. in Computer Science from Princeton University in 2010 under the supervision of Sanjeev Arora. His doctoral dissertation, "On the Complexity of Unique Games and Graph Expansion," received an Honorable Mention for the ACM Doctoral Dissertation Award in 2011. Steurer's research focuses on algorithm design through mathematical programming relaxations, particularly semidefinite programming and the sum-of-squares method. His work addresses computational complexity of high-dimensional estimation problems including tensor decomposition, clustering, Gaussian mixture models, and stochastic block models. He also investigates algorithmic aspects of robustness and differential privacy, especially for estimation tasks. His theoretical contributions have practical implications for machine learning and data analysis in the presence of noise and adversarial conditions. Analysis of Steurer's recent publications reveals a consistent trajectory in advancing the sum-of-squares framework for high-dimensional estimation and robust statistics. His work demonstrates how sophisticated mathematical programming techniques can provide certifiable guarantees for challenging problems in machine learning. The research spans theoretical foundations while maintaining relevance to practical applications, particularly in scenarios involving corrupted or noisy data. A notable trend is the extension of sum-of-squares methods to handle increasingly complex statistical models while providing rigorous performance guarantees. ERC Consolidator Grant (2019) Michael and Sheila Held prize (2018, with Raghavendra) Invited speaker at International Congress of Mathematicians (2018) STOC Best Paper Award (2015) Alfred P. Sloan Research Fellowship (2014) NSF CAREER Award (2014) Microsoft Research Faculty Fellowship (2014) FOCS Best Paper Award (2010) Steurer has advised numerous PhD students and postdocs who have gone on to prominent positions, including Sam Hopkins (now faculty at MIT), Jonathan Shi, and Aaron Potechin. His research is supported by multiple prestigious grants, including an ERC Consolidator Grant, an NSF CAREER Award, and a Microsoft Research Faculty Fellowship. He has served on numerous program committees for top theoretical computer science conferences and has been recognized for his service to the community through roles such as internal member of the Scientific Advisory Committee of the Institute for Theoretical Studies at ETH Zurich. As head of the Institute for Theoretical Computer Science at ETH Zurich, Steurer leads a research group focused on advancing the theoretical foundations of computer science with particular emphasis on mathematical optimization methods. His team explores the boundaries of what can be efficiently computed in high-dimensional settings, with applications spanning machine learning, statistics, and data science. The group maintains strong connections with both theoretical and applied researchers across ETH and the broader academic community.
Professor Guoyin Li is a faculty member at the School of Mathematics & Statistics , University of New South Wales (UNSW Sydney). He holds a Ph.D. from The Chinese University of Hong Kong (2007) and has been at UNSW since 2011, currently serving as Professor and Research Director. Research Interests His work spans optimization , variational analysis , and multilinear algebra , with applications in robust optimization , structural engineering , and machine learning . He specializes in nonconvex nonsmooth optimization , tensor eigenvalue problems , and exact semi-definite programming relaxations . Recent Publications His articles focus on robust optimization for structural design, nonlinear approximation techniques, and conic programming for uncertain data. Key journals include Foundations of Computational Mathematics , Mathematical Programming , and Computer Methods in Applied Mechanics and Engineering . Awards and Grants Fellow of the Australian Mathematical Society (2023) 2022 AustMS Medal 2024 Marguerite Frank Award ARC Discovery Grants (2021-2023, 2025-2027) ARC Research Hub Project (2017-2021) Professional Roles He serves on editorial boards of SIAM Journal on Optimization , Optimization Letters , and Journal of Optimization Theory and Applications , and has delivered plenary lectures at international conferences in Austria, Spain, and Canada.
David P. Woodruff is a Professor in the Department of Computer Science at Carnegie Mellon University, part of the Theory Group within the School of Computer Science. He is actively involved in academic leadership roles, including chairing the CATCS (Conference on Theoretical Computer Science) and serving as PC chair for SODA 2024 and ICALP 2022. His research focuses on algorithms, data streams, machine learning, numerical linear algebra, sketching, and sparse recovery. He has been recognized with awards such as the Herbert Simon Award for teaching and the PODS Best Paper Award. Woodruff has advised numerous students and postdocs, including notable scholars like Ainesh Bakshi, Rajesh Jayaram, and Hongyang Zhang. His work often addresses foundational challenges in theoretical computer science, with contributions to distributed computing, streaming algorithms, and privacy-preserving techniques. He has published extensively in top conferences like NeurIPS, ICML, FOCS, and STOC, covering topics ranging from low-rank approximation to adversarial robustness in data streams. His teaching includes courses like Algorithms for Big Data and core algorithms courses, reflecting his commitment to both research and education. Collaborations span academia and industry, with applications in genomics and secure computation. Woodruff is a key contributor to the Foundations of Data Science program at the Simons Institute.
Professor Saman Amarasinghe is a full Professor in the Department of Electrical Engineering and Computer Science (EECS) at the Massachusetts Institute of Technology (MIT), and Principal Investigator at MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL). He leads the Commit compiler research group, which focuses on programming languages and compilers that maximize application performance on modern computing platforms. His work spans multiple academic departments and research centers, with strong affiliations to both MIT's School of Engineering and CSAIL. Professor Amarasinghe's research interests center around high-performance domain-specific languages and compiler technology . His work combines language design with sophisticated compilation techniques to deliver unprecedented performance for targeted application domains. His research spans multiple areas including image processing (Halide), sparse tensor algebra (TACO), graph analytics (GraphIt), stream computations (StreamIt), and bioinformatics (Seq). A significant thread throughout his work is the application of machine learning for compiler optimizations, from Meta optimization in 2003 to the OpenTuner autotuner framework. Analysis of Professor Amarasinghe's recent publications reveals a strong focus on sparse computing , compiler vectorization , and domain-specific language implementation . His work consistently bridges theoretical compiler concepts with practical performance gains across diverse application domains. The progression from earlier work on StreamIt and Halide to more recent projects like GraphIt and TACO shows an evolution toward more specialized, high-performance DSLs targeting specific computational patterns. His 2020-2025 publications particularly emphasize sparse tensor operations, GPU acceleration, and machine learning integration with compiler technology. ACM Fellow (2019) Professor Amarasinghe has made significant contributions to academic entrepreneurship and student development. He founded Determina, Inc. (acquired by VMware) based on security research from his MIT lab and co-founded Lanka Internet Services, Ltd., Sri Lanka's first ISP. As faculty director of MIT Global Startup Labs, his programs across 17 countries have helped create over 20 successful startups. His teaching includes the popular Performance Engineering of Software Systems (6.172) course with Professor Charles Leiserson, as well as innovative project-based courses like the Open Source Software Project Lab and Bring Your Own Software Project Lab. His educational approach emphasizes hands-on experience with compiler and language design concepts. Professor Amarasinghe leads the Commit compiler research group at MIT CSAIL, which has produced numerous influential domain-specific languages and compilers including Halide, TACO, Simit, StreamIt, and GraphIt. The lab maintains strong industry connections through projects like OpenTuner and Determina, and collaborates with researchers worldwide on compiler technology. The group's work spans both theoretical compiler research and practical implementation, with a consistent focus on bridging the performance gap between high-level programming abstractions and hardware capabilities.
P. (Saday) Sadayappan is a Professor in the School of Computing at the University of Utah. He serves as a lead researcher in high-performance computing, with a focus on compiler optimization and algorithm-architecture co-design. His current projects include NIH SBIR Phase 2 funding for large-scale image analysis and NSF grants for tensor applications and cyber-infrastructure for AI. Research Interests : Compiler Optimization for High Performance Computing Optimization of Sparse/Dense Matrix/Tensor Computations Scalable Machine Learning Algorithm-Architecture Co-Design Optimization Research Trends in Publications : His work emphasizes optimizing computational workflows for emerging hardware architectures, with a focus on accelerating machine learning and scientific computing through compiler-level innovations. Recent trends include co-design for CNNs, sparse matrix optimizations, and distributed algorithms. Scientific Awards : ACM SIGPLAN Most Influential PLDI Paper Award (2018) Grants & Projects : NSF (2022–2027): Comprehensive Framework for Tensor Applications NSF AI Institute ICICLE (2021–2026): Cyber-infrastructure for environmental AI NIH SBIR (2023–2025): Next-gen machine learning for image analysis Labs & Teams : Collaborates with institutions like Ohio State University and RNET Technologies on projects involving parallel computing, sparse algorithms, and compiler design.
