Find Academic Supervisors

Mark SchmidtView profile

Professor

• Machine Learning
• Artificial Intelligence
• Mathematical Optimization
• +6 more

Mark Schmidt is a Professor in the Department of Computer Science at the University of British Columbia, Faculty of Science. He holds the Canada Research Chair in Large-Scale Machine Learning and is a Canada CIFAR AI Chair at the Alberta Machine Intelligence Institute. His research spans multiple centers including CAIDA (Centre for Artificial Intelligence Decision-making and Action), the Data Science Institute, and the Machine Intelligence Learning Discovery (MILD) group. Dr. Schmidt's educational background includes a Ph.D. from UBC (2005-2010), an M.Sc. from the University of Alberta (2003-2005), and a B.Sc. from the University of Alberta (2000-2003). His academic career progressed from Postdoc positions at Simon Fraser University, Ecole Normale Superieure, and UBC to Assistant Professor (2014-2019), Associate Professor (2019-2024), and current Professor (2024-present) at UBC. His research focuses on machine learning optimization, with particular emphasis on improving numerical algorithms for large-scale machine learning applications. His work bridges theoretical optimization and practical applications across computer vision, natural language processing, and reinforcement learning. He has developed numerous optimization techniques including variants of stochastic gradient methods, coordinate descent algorithms, and natural gradient approaches. Analysis of his recent publications reveals a strong focus on optimization for over-parameterized models, particularly transformers and large language models. His work addresses critical challenges in step-size selection, convergence guarantees, and efficient implementation of optimization algorithms. His research has significant implications for training deep neural networks more effectively and understanding why certain optimization methods outperform others in practice. Among his notable awards are the Dorothy Killam Fellowship (2025), Arthur B. McDonald Fellowship (2024), Sloan Research Fellowship (2017), and multiple UBC teaching awards. He has also received the Lagrange Prize in Continuous Optimization and Best Paper Award at AISTATS 2021. Dr. Schmidt actively supervises numerous graduate students, with over 40 PhD and Master's students listed as current or alumni members of his research group. His laboratory maintains strong connections with industry partners, with many alumni securing positions at leading AI companies including Google, Amazon, Meta, and Microsoft.

Amitabha BagchiView profile

Professor

• Data Algorithmics
• Probability and Networks
• Theoretical Computer Science
• +7 more

Amitabha Bagchi is a Professor in the Department of Computer Science and Engineering at IIT Delhi. His research spans data algorithmics, probability, networks, and theoretical computer science, with applications in distributed systems, social networks, and AI-driven platforms. He has published extensively in leading venues such as SIGMOD, VLDB, ICDE, AAAI, and KDD, often collaborating with students and researchers on problems involving graph algorithms, fairness, and large-scale data analysis. Research Interests: His primary research interests include Data Algorithmics, Probability and Networks, Theoretical Computer Science, Distributed Algorithms, Graph Algorithms, and Machine Learning Theory. He investigates algorithmic foundations for real-world problems such as food delivery optimization, social network analysis, and efficient data structures for streaming and large graphs. Publication Trends: Recent publications focus on fairness in gig economy platforms, efficient solvers for graph Laplacians, generalization in neural networks, and temporal graph querying. His work combines theoretical rigor with practical impact, often involving GPU acceleration, distributed computing, and data-aware algorithm design. Scientific Service: Editor, Algorithms (2020–present) Editor, Journal of Discrete Algorithms , Elsevier (2006–2018) Guest Editor, special issue on Algorithms for Shortest Paths in Dynamic and Evolving Networks , Algorithms (2021) Volume Editor for proceedings of ESA, ATMOS, COCOON, and others Conference Leadership: He has served on numerous program committees and as chair for conferences including ESA (Engineering Track, 2016), ATMOS (2020), and ICALP (2019). His involvement spans algorithmic engineering, transportation optimization, and theoretical computer science forums. Teaching: He currently teaches COL863: Special Topics in Theoretical Computer Science on concentration inequalities and their applications. He has previously taught advanced courses in algorithms and data structures.

Mark CrowleyView profile

Associate Professor

• Reinforcement Learning
• Deep Learning
• Manifold Learning
• +12 more

Mark Crowley is an Associate Professor in the Department of Electrical and Computer Engineering at the University of Waterloo, with a cross-appointment in the Cheriton School of Computer Science. He is a member of the Waterloo Artificial Intelligence Institute (WAII) and the Waterloo Institute for Complexity and Innovation (WICI), and serves as National Secretary of the Canadian Artificial Intelligence Association (CAIAC). His educational background includes a Ph.D. and M.Sc. in Computer Science from the University of British Columbia, where he worked in the Laboratory for Computational Intelligence, and a B.A. in Computer Science from York University. He completed a postdoctoral fellowship at Oregon State University working with Tom Dietterich's machine learning group. Crowley's research focuses on developing dependable and transparent algorithms to augment human decision-making in complex domains with multiple agents, spatial structure, or uncertainty. His work spans Reinforcement Learning , Deep Learning , Ensemble Methods , and Manifold Learning . He frequently collaborates with researchers in applied fields including Computational Sustainability, Sustainable Forest Management, Autonomous Driving, Medical Imaging, and Material Design. His research is motivated by both theoretical opportunities and real-world challenges such as forest fire management, automotive applications, and medical imaging. His recent publications demonstrate a strong focus on addressing challenges in reinforcement learning, particularly around observation costs, multi-agent systems, and causal representation learning. His work on ChemGymRL provides a significant contribution to digital chemistry and material design through reinforcement learning frameworks. The textbook Elements of Dimensionality Reduction and Manifold Learning represents a major contribution to the theoretical foundations of machine learning. Crowley actively supervises graduate students, with recent thesis completions including Shayan Shirahmadi Gale Bagi (PhD, Feb 2025) and Oleksandra Nahorna (MASc, Dec 2024). His lab, UWECEML (Waterloo ECE Machine Learning Lab), focuses on developing new algorithms at the intersection of Machine Learning, Optimization, and Probabilistic Modeling. He teaches courses including ECE 457C (Reinforcement Learning), ECE 657A (Data & Knowledge Modelling & Analysis), and ECE 457B (Fundamentals of Computational Intelligence). His blog Computationally Thinking explores AI, machine learning, and the societal impact of these technologies.

• Optimization
• Complex Networks
• Statistical Inference
• +3 more

Stefano NASINI is an Associate Professor at the University of Lille 3, specializing in Quantitative Methods within the Economics and Mathematics Sciences. He holds a HDR (Habilitation à Diriger des Recherches) from the University of Lille 3 (2021), a Ph.D. in Statistics and Operations Research from the Polytechnic University of Catalonia (2015), and a Master in Statistics (2011). His research focuses on optimization, complex networks, statistical inference, and microeconomic applications. He has held academic positions including a post-doctoral role at IESE Business School (2014–2016) and a visiting researcher role at the University of Lisbon (2014). His work spans scheduling optimization, network analysis, financial contagion modeling, and energy system planning. Key contributions include specialized algorithms for large-scale optimization problems and frameworks for decentralized portfolio management. He is a member of the LEM research group and teaches courses in optimization, econometrics, and social network analysis at the Grande École and MSc levels. Publications highlight interdisciplinary applications, including network-based diffusion models, multi-market financial strategies, and dynamic choice analysis. His research bridges theoretical advancements in operations research with practical challenges in economics, energy, and transportation systems. No scientific awards are explicitly listed in the provided materials. His advising roles and grants are not detailed here, but his extensive publication record reflects active collaboration within academic and applied domains.

Florian ShkurtiView profile

Assistant Professor

• Robotics
• Machine Learning
• Computer Vision
• +6 more

Florian Shkurti is an Assistant Professor in the Department of Computer Science at the University of Toronto Mississauga (UTM), affiliated with the UofT Robotics Institute, Vector Institute, and Acceleration Consortium. His research focuses on robotics, machine learning, and computer vision, emphasizing safe and effective autonomous systems in dynamic environments. He directs the Robot Vision and Learning (RVL) lab, exploring areas like environmental monitoring, autonomous navigation, and mobile manipulation. Research Interests: His work spans robotics, machine learning, and computer vision. Key areas include robot perception, planning under uncertainty, safe exploration, imitation learning, and applications in field robotics, autonomous vehicles, and chemistry lab automation. He develops methods enabling robots to perceive, reason, and act safely in collaboration with humans. Publications: Recent work includes advancements in safe multitask learning, interactive crowd navigation, and diffusion models for trajectory planning. His research bridges theoretical foundations with real-world applications in environmental science and autonomous systems. Affiliations: Faculty Member, UofT Robotics Institute; Faculty Affiliate, Vector Institute; Faculty Member, Acceleration Consortium. He also holds positions at UTM's Mathematical & Computational Sciences department. Teaching: Courses include Imitation Learning for Robotics, Neural Networks, and Mobile Robotics. He emphasizes hands-on experience with autonomous systems through projects involving RC cars and simulation tools. Labs & Teams: Leads the RVL lab, collaborating on projects like RoboCulture (automated biological experimentation) and SICNav (safe crowd navigation systems). The lab focuses on cross-disciplinary robotics solutions for real-world challenges.

Westley WeimerView profile

Professor

• Automated Program Repair
• Software Engineering
• Programming Languages
• +18 more

Westley Weimer is a Professor in the Department of Electrical Engineering and Computer Science (EECS) at the University of Michigan, College of Engineering. He teaches advanced courses such as EECS 590 (Advanced Programming Languages) and EECS 481 (Software Engineering), and has previously taught at the University of Virginia. His research integrates software engineering, programming languages, and cognitive science, focusing on automated program repair, program analysis, and the neuroscience of code comprehension. University: University of Michigan School: College of Engineering Department: Department of Electrical Engineering and Computer Science Academic Rank: Professor His research interests include automated program repair (e.g., GenProg), software quality, cognitive modeling of programming, neuroimaging studies of code review, and the application of medical imaging to software engineering. He explores deep questions at the intersection of consciousness, time, and computation, advocating for interdisciplinary approaches to understanding the mind through programming behavior. The most recent publications reflect a trend toward empirical and cognitive studies in software engineering, combining automated repair with human factors, neuroimaging (fMRI, TMS), and real-world software challenges. Themes include bias in code review, programming under cognitive influences, and the neurological basis of code comprehension. His work increasingly bridges computer science with psychology, neuroscience, and social science. Scientific awards include multiple Distinguished Paper Awards at ICSE, FSE, and ESEC/FSE, Best Paper and Runner-up awards, and several 10-Year Most Influential Paper Awards from ASE, GECCO, POPL, and ASPLOS, recognizing the lasting impact of his contributions to automated software repair and program analysis. He has advised numerous PhD and Master’s students, many of whom have gone on to faculty positions or industry research roles. He contributes to academic service through organizing diversity and inclusion initiatives, maintaining graduate career resources, and promoting ethical and inclusive practices in computing. He leads a vibrant research group focused on improving software quality through both technical and human-centered innovations, with ongoing projects in automated repair, cognitive modeling, and secure systems.

Mathias NiepertView profile

Professor

• Artificial Intelligence
• Machine Learning
• Graph Neural Networks
• +5 more

Mathias Niepert is a Professor at the Institute for Artificial Intelligence within the Faculty of Computer Science, Electrical Engineering and Information Technology at the University of Stuttgart. His research focuses on advancing machine learning techniques with applications in scientific computing, graph neural networks, and medical imaging. He is particularly known for contributions to physics-informed neural networks, equivariant models, and graph learning frameworks. Key research areas include: Scientific Machine Learning for PDEs and molecular modeling Graph neural networks and their theoretical limitations Medical vision-language models and multimodal learning Efficient neural network architectures (transformers, FNOs) Domain knowledge integration in deep learning His work often bridges theoretical foundations with practical applications, as evidenced by extensive publications (2018–2025) on topics like adaptive message passing, equivariant networks, and medical imaging systems. He has contributed to benchmark development through initiatives like PDEBench and pioneered methods for equivariant diffusion models and molecular representation learning. His current projects emphasize: Improving generalization in Fourier Neural Operators Addressing oversmoothing in graph networks Combining physics principles with neural architectures Medical AI applications through multimodal fusion

Carey E. PriebeView profile

Professor

• computational statistics
• statistical pattern recognition
• statistical image analysis
• +5 more

Carey E. Priebe is a Professor in the Department of Applied Mathematics and Statistics at the Whiting School of Engineering, Johns Hopkins University. He maintains strong affiliations with multiple research centers including the Johns Hopkins University Center for Imaging Science, the Mathematical Institute for Data Science, and the Human Language Technology Center of Excellence. His academic career spans several decades with significant contributions to statistical methodology and theory. Dr. Priebe's research focuses on computational statistics, statistical pattern recognition, and statistical inference for high-dimensional and graph data. His work bridges theoretical statistics with practical applications in areas such as brain connectome mapping, network analysis, and image processing. He has made significant contributions to spectral graph theory, graph matching, and vertex nomination, with applications ranging from neuroscience to national security. His publication record demonstrates consistent contributions to statistical methodology, with a notable emphasis on graph-based statistical methods. His research trajectory shows increasing focus on network data analysis, particularly in the last decade, with applications to brain mapping and connectome analysis as evidenced by his NSF BRAIN Initiative grant and Nature publication. 2013 Erskine Fellow (University of Canterbury) 2011 McDonald Award for Excellence in Mentoring and Advising 2010 ASA SDNS Distinguished Achievement Award 2009 Erskine Fellow (University of Canterbury) 2008 National Security Science and Engineering Faculty Fellow 2008 Pond Award for Excellence in Teaching NSF BRAIN EAGER grant recipient (2014) Professor Priebe has supervised an extensive number of doctoral students whose work spans statistical methodology, network analysis, and machine learning. His students have secured positions at prestigious institutions including academia (University of Wisconsin, Boston University), government research labs, and major technology companies (Microsoft, Facebook, Amazon). His research has been supported by significant grants from NSF, DARPA, and other agencies focused on national security applications and fundamental statistical methodology development. He maintains active collaborations across multiple disciplines and institutions, as evidenced by his numerous conference presentations and visiting appointments including at The Alan Turing Institute and The Isaac Newton Institute. His work bridges theoretical statistics with practical applications in neuroscience, security, and data science.

Salim El RouayhebView profile

Associate Professor

• Information Theory
• Coding Theory
• Information-Theoretic Security
• +4 more

Salim El Rouayheb is an Associate Professor in the Department of Electrical and Computer Engineering at Rutgers University. He leads the Coding and Securing Information (CSI) Lab, which focuses on information-theoretic security and privacy in distributed systems. His research spans multiple areas including secure machine learning, private information retrieval, and data synchronization. Dr. El Rouayheb received his Ph.D. in Electrical Engineering from Texas A&M University in 2009. Prior to joining Rutgers, he was an Assistant Professor at the Illinois Institute of Technology (2013-2017), a Research Scholar at Princeton University (2012-2013), and a Postdoctoral Researcher at UC Berkeley (2010-2011). His research interests focus on information-theoretic security in distributed systems, private information retrieval and search, secure machine learning algorithms, and data synchronization in distributed systems. He has made significant contributions to developing frameworks that provide information-theoretic privacy guarantees in various contexts including federated learning, genomic data analysis, and decentralized networks. His work often bridges theoretical foundations with practical applications, particularly in the areas of secure distributed computing and privacy-preserving algorithms. His recent publications demonstrate a strong trend toward applying information-theoretic principles to address privacy and security challenges in machine learning systems, particularly in federated and decentralized settings. Many of his papers explore random walk approaches for decentralized learning, secure matrix multiplication techniques, and privacy mechanisms that can be toggled "on and off" based on correlation patterns in data. His work spans both theoretical contributions in information theory and practical implementations for real-world systems. Dr. El Rouayheb has received several prestigious awards including the NSF CAREER Award (2016), Google Faculty Research Award (2018), and the Rutgers University Walter Tyson Junior Faculty Chair (2019). He has successfully secured multiple research grants including NSF SaTC, NSF CAREER, Google Faculty Research Awards, and Army Research Lab funding. His lab, the Coding and Securing Information (CSI) Lab, currently includes postdoc Xingran Chen, PhD student Zonghong Liu, and undergraduate researchers. The CSI Lab maintains an active research agenda with regular publications in top-tier venues and hosts the Shannon Channel, a series of online talks related to information theory. Dr. El Rouayheb is also involved in organizing workshops on coding theory and information security.

Mark CrowleyView profile

Associate Professor

• Reinforcement Learning
• Deep Learning
• Ensemble Methods
• +6 more

Mark Crowley is an Associate Professor in the Department of Electrical and Computer Engineering at the University of Waterloo , with a cross-appointment in the Cheriton School of Computer Science . He is actively involved in the Waterloo Artificial Intelligence Institute (WAII) , the Waterloo Institute for Complexity and Innovation (WICI) , and serves as National Secretary for the Canadian Artificial Intelligence Association (CAIAC) , coordinating the Canadian Conference on AI . Research interests span the theoretical and applied aspects of Reinforcement Learning , Deep Learning , Manifold Learning , and Ensemble Methods . His work addresses challenges in domains with spatial dynamics, multi-agent systems, and uncertainty, particularly in Computational Sustainability (forest fire management, sustainable forestry), Autonomous Driving , Medical Imaging , and Material Design . Recent research focuses on integrating causal modeling with generative representation learning to improve out-of-distribution robustness in motion forecasting applications. Key publications include foundational work on ChemGymRL environments for safe chemical process reinforcement learning, Generative Causal Representation Learning for robust forecasting, and collaborative work on multi-advisor reinforcement learning in multi-agent settings. He co-authored a textbook Elements of Dimensionality Reduction and Manifold Learning (Springer, 2023) with Prof. Ali Ghodsi and Prof. Fakhri Karray. Teaching includes graduate and undergraduate courses in Algorithm Design , Computational Intelligence , Reinforcement Learning , and Data Modeling at the University of Waterloo since 2018. His research group has produced several notable graduates including Benyamin Ghojogh (2021), who continued as a postdoc until 2022.

Yuanzhu ChenView profile

Professor

• Computer Networking
• Mobile Computing
• Complex Networks
• +5 more

Yuanzhu Chen is a Professor in the School of Computing at Queen’s University, affiliated with the Faculty of Arts and Science. He previously served as Professor and Department Head at Memorial University of Newfoundland (2005–2021). His research focuses on computer networking, mobile computing, complex networks, and applied machine learning, emphasizing wireless innovation beyond traditional wired systems. He holds a PhD from Simon Fraser University (2004) and a B.Sc. from Peking University (1999). Education: PhD in Computing Science (Simon Fraser University, 2004); B.Sc. in Computer Science (Peking University, 1999). Earlier roles include Post-doctoral Researcher at Simon Fraser University (2004–2005) and leadership positions at Memorial University, including Department Head (2019–2021). Research Interests: Network Coding and Opportunistic Routing Mobile and Wireless Network Protocols Complex Network Analysis Machine Learning Applications Indoor Positioning Systems Social Network Dynamics Selected Awards: Recipient of Queen’s University President's Award for Distinguished Teaching. Lab Affiliation: Director of the Wireless Networking and Mobile Computing Lab (WineMocol). Active in collaborative projects involving smartphone sensors, community-based environmental monitoring, and stock market prediction using web data.

Brian C. WilliamsView profile

Professor

• Artificial Intelligence
• Robotics
• Autonomous Systems
• +4 more

Professor Brian C. Williams is a leading academic at the Massachusetts Institute of Technology (MIT) , holding the position of Professor of Aeronautics and Astronautics and directing the Autonomous Systems Laboratory (ASL) . He is also a member of the Computer Science and Artificial Intelligence Laboratory (CSAIL) and the Space Systems Laboratory (SSL) . His work focuses on advancing model-based autonomy, enabling robots to operate independently in extreme environments such as space, underwater, and urban traffic. His contact details include email williams@csail.mit.edu and phone 253-2739 . Education : S.B., S.M., and Ph.D. in Computer Science and Electrical Engineering from MIT (1989). Williams' research spans risk-bounded decision making , collaborative robotics , and neural-symbolic learning . He has pioneered systems like Remote Agent, which demonstrated autonomous self-repair in NASA's Deep Space One mission, and Geordi, a risk-aware driver assistant. His work integrates symbolic reasoning, probabilistic methods, and machine learning to create resilient robotic systems for space exploration, manufacturing, and transportation. The most recent publications highlight trends in risk-aware planning , multi-agent robotics , and stochastic control . Key innovations include tube-based trajectory optimization, conflict-directed task allocation, and theory-of-mind guided interventions, reflecting his focus on robustness in uncertain environments. Scientific Awards : NASA Space Act Award (1999), AAAI Fellow, and multiple best paper awards from AAAI, IJCAI, ICAPS, CDC, HRI, and ECAI. Williams leads the Model-Based Embedded and Robotics Systems Group at CSAIL, collaborating with institutions like NASA's Jet Propulsion Laboratory. His projects include ASIST (Artificial Social Intelligence for Teams), RADMAX (Risk and Deadline Aware Planning), and Uhura (risk-aware personal assistants), addressing challenges in health care, manufacturing, and defense applications.

Sebastijan DumancicView profile

Assistant Professor

• Program Synthesis
• Probabilistic Programming
• Neuro-Symbolic AI
• +2 more

Sebastijan Dumancic is an Assistant Professor at Delft University of Technology, focusing on neuro-symbolic AI through program synthesis and probabilistic programming. He leads the RAIL lab and collaborates with institutions like Harvard, MIT, and CNRS. His research bridges symbolic AI and machine learning, applying program synthesis to scientific discovery, transportation, and robotics. He holds an FWO-funded PhD from KU Leuven and has participated in initiatives like ELLIS and the Symbolic Computation and Machine Learning Initiative. Program synthesis Probabilistic programming Neuro-symbolic AI Constraint-based learning His recent articles highlight advancements in program synthesis, neuro-symbolic integration, and constraint satisfaction. Projects like Find2Fix and Intelligent Greenhouse Horticulture (funded by NWO) demonstrate practical applications. ELLIS Membership University Teaching Qualification He supervises numerous MSc and PhD students in projects involving logic programming, program synthesis, and probabilistic modeling. Active in workshops and symposia, he contributes to neuro-symbolic AI and scientific discovery.

Todd D. MurpheyView profile

Professor

• Computational Dynamics and Control
• Human-Swarm Interaction
• Algorithmic Matter
• +10 more

Dr. Todd D. Murphey is a Professor of Mechanical Engineering at Northwestern University's Robert R. McCormick School of Engineering and Applied Science. He serves as Director of Transformative Research and Director of the Master of Science in Robotics Program at Northwestern, leading initiatives in computational dynamics, control systems, and robotics. His work bridges engineering, neuroscience, and biomedical applications, with a focus on developing systems that interact effectively with humans and their environments. Dr. Murphey received his Ph.D. in Control and Dynamical Systems from the California Institute of Technology in 2002, with a thesis titled "Control of Multiple Model Systems." Prior to that, he earned a B.S. in Mathematics, summa cum laude, from the University of Arizona in 1997. Dr. Murphey's research centers on computational methods in dynamics and control, with applications spanning neuroscience, health science, robotics, and automation. His work in the Interactive & Emergent Autonomy Lab focuses on computational models of embedded control, biomechanical simulation, dynamic exploration, and hybrid control. The group develops mathematical approaches that lead to orders of magnitude improvement in computational efficiency for real-time implementation. Key application areas include assistive exoskeleton control, stabilization of energy networks, bio-inspired active sensing, entertainment robots, robotic exploration, and software-enabled stroke rehabilitation. Analysis of Dr. Murphey's recent publications reveals a strong emphasis on human-swarm interaction, algorithmic matter, and control of cyber-physical systems in uncertain environments. His work increasingly integrates information theory with physical systems, exploring how both autonomous and biological systems interact with environments to learn and improve behaviors. Recent trends show growing applications in rehabilitation technology, with particular focus on human-machine interaction in biomedical devices and embodied intelligence. Dr. Murphey has received numerous honors and awards for his contributions to robotics and engineering: Named Director of Transformative Research at Northwestern University (2025) Appointed IEEE Robotics and Automation Society Vice President of Publication Activities (2022) Co-recipient of Best Paper Award for IEEE Transactions on Robotics (2020) Appointed to Air Force Scientific Advisory Board (2019) Recipient of ABB Best Student Paper Award for CPL-SLAM research (2019) Cole-Higgins Award from Northwestern Engineering (2015) Dr. Murphey has supervised numerous graduate students including Taosha Fan, Giorgos Mamakoukas, and Ian Abraham, with research spanning robotic exploration using electrosense and mechanical contact, human-in-the-loop control, and shared control for rehabilitation devices. His lab has secured significant funding from the National Science Foundation, DARPA, and industry partners including Siemens and Ekso Bionics, supporting research in algorithmic matter, emergent behavior, and human-swarm collaboration. The Interactive & Emergent Autonomy Lab, led by Dr. Murphey, investigates how both autonomous systems and biological systems interact with their environments to learn and improve behaviors. Current projects include active learning and data-driven control, active perception in human-swarm collaboration, algorithmic matter and emergent computation, control for nonlinear and hybrid systems, cyber physical systems in uncertain environments, harmonious navigation in human crowds, information maximizing clinical diagnostics, reactive learning in underwater exploration, robot-assisted rehabilitation, and software-enabled biomedical devices. The lab collaborates with researchers across Northwestern and institutions including Georgia Tech, MIT, and industry partners.

Albert S. BerahasView profile

Assistant Professor

• Nonlinear Optimization
• Machine Learning Optimization
• Stochastic Optimization
• +5 more

Albert S. Berahas is an Assistant Professor in the Department of Industrial and Operations Engineering at the University of Michigan's College of Engineering. He joined the university in 2020 after completing postdoctoral positions at Lehigh University (2018-2020) and Northwestern University (2018). He holds a PhD in Engineering Sciences and Applied Mathematics from Northwestern University (2018), an MS in Applied Mathematics from Northwestern (2012), and a BSE in Operations Research and Industrial Engineering from Cornell University (2009). His research focuses on designing, developing, analyzing, and implementing algorithms for solving large-scale nonlinear optimization problems. His work spans multiple sub-fields including constrained optimization, optimization for machine learning, stochastic optimization, derivative-free optimization, and decentralized optimization. He is affiliated with the Michigan Institute for Data Science (MIDAS), the Michigan Institute for Computational Discovery and Engineering (MICDE), and the Michigan Center for Applied and Interdisciplinary Mathematics (MCAIM). Berahas has received numerous honors including the Charles Broyden Prize (2025), the Air Force Office of Scientific Research Young Investigator Program award (2025), the IISE Operations Research Division Teaching Award (2024), and the North Campus Dean's MLK Spirit Award for Community Building & Impact (2024). His recent publications demonstrate strong activity in developing novel optimization frameworks with theoretical guarantees for challenging problem settings. His research has been supported by significant grants including from the Office of Naval Research (ONR) and the Air Force Office of Scientific Research. He actively mentors PhD students and has successfully advised Jiahao Shi, who defended his dissertation in March 2025 and joined Amazon. Berahas is also engaged in community outreach, particularly through initiatives like Engage Detroit that aim to empower Detroit's next generation of engineers.