Find Academic Supervisors

• Machine Learning
• Bayesian Nonparametrics
• Statistical Theory
• +3 more

Peter Orbanz is a Professor of Machine Learning at the Gatsby Computational Neuroscience Unit, University College London (UCL). He previously held roles as Assistant and Associate Professor in the Department of Statistics at Columbia University, and completed postdoctoral research at the University of Cambridge under Zoubin Ghahramani. His research focuses on machine learning, Bayesian nonparametrics, symmetry in statistical models, and dynamics. He has spent sabbaticals at the Isaac Newton Institute, Microsoft Research New England, and UC Berkeley. Orbanz earned his PhD in 2008 from ETH Zurich under Joachim M. Buhmann, with a thesis on Bayesian nonparametric models. His work bridges theoretical foundations of machine learning with practical applications, emphasizing symmetry properties and scalable inference methods. Current research includes causal uncertainty quantification, Gaussian process dynamics, and equivariant neural networks. He teaches advanced courses on probabilistic learning, Bayesian nonparametrics, and statistical theory at UCL and Columbia. His research group at UCL includes PhD students Hugh Dance, Kevin H. Huang, Vasco Portilheiro, and Vince Velkey. Former advisees include Morgane Austern (Harvard), Benjamin Bloem-Reddy (UBC), and Victor Veitch (Chicago). Orbanz has authored influential papers on topics such as exchangeable random structures, Gaussian universality, and scalable MCMC methods. His work often integrates concepts from probability theory, statistical physics, and computational neuroscience. He maintains active collaborations with institutions worldwide and contributes to open-source tools for Bayesian modeling.

Cristian ProistosescuView profile

Assistant Professor

• Climate Dynamics
• Climate Variability and Change
• Cloud Physics and Radiative Processes
• +5 more

Cristian Proistosescu is an Assistant Professor in the Department of Climate, Meteorology & Atmospheric Sciences at the University of Illinois at Urbana-Champaign, part of the School of Earth, Society & Environment. He holds a B.A. in Physics from Princeton University (2009) and a Ph.D. in Earth and Planetary Sciences from Harvard University (2017). His research focuses on coupled atmosphere-ocean dynamics, climate sensitivity, stochastic climate variability, and the application of statistical inference and machine learning to earth sciences. Proistosescu’s work addresses critical questions such as the drivers of climate variability (e.g., low-cloud feedbacks, western Pacific warming patterns), the integration of paleoclimate data to constrain future projections, and the economic implications of climate risk. His research group, Climate Dynamics and Data Science (CDDS) @ UIUC, explores interdisciplinary topics including climate economics and exploratory projects like seafloor bathymetry analysis. Highlighted Contributions: Co-authored influential studies on climate sensitivity (e.g., Reviews of Geophysics 2020), and developed frameworks for understanding feedbacks via Green’s function approaches. Courses Taught: Climate and Global Change (ATMS 140), Climate Dynamics (ATMS 507), and Statistical Inference for Earth Sciences (GEOL 593/ATMS 597). Recent Activities: Launched new projects on carbon dioxide as a risky asset and climate-economic linkages, with Ph.D. students Maile and Pappu joining in 2022. His team collaborates across disciplines, emphasizing data-driven approaches to address climate challenges. Proistosescu is affiliated with the Department of Geology and actively engages with international conferences (e.g., AGU 2022 presentations).

• Aerospace Engineering
• Materials Science
• Plasma Physics
• +3 more

Nagi Mansour is an Adjunct Research Professor at the University of Illinois, specializing in advanced aerospace materials and thermal protection systems. His research focuses on understanding material behavior under extreme aerothermal conditions, including high-enthalpy plasma environments, hypersonic flow, and space entry scenarios. He leads computational and experimental studies on ablation mechanisms, material pyrolysis, and the development of predictive models for thermal protection systems (TPS). Key areas of expertise include: (1) Material Response Modeling of carbon-based and silicone-coated composites, (2) Plasma Environment Simulation using tools like PuMA and PATO, and (3) High-Fidelity Numerical Analysis of arc-jet and plasmatron test facilities. His work integrates multi-scale analysis from microstructure characterization (via X-ray tomography) to macroscopic aerothermal performance prediction. Recent contributions include advancements in porous microstructure analysis software (PuMA), unified flow-material approaches for ablation modeling, and experimental validation of hypersonic TPS materials like PICA-NuSil. His research supports NASA missions and future planetary entry probe designs, addressing challenges such as Mars atmospheric entry and Titan exploration. Collaborative projects involve institutions like the VKI Plasmatron facility and NASA's Hypersonic Materials Environmental Test System (HyMETS). Current efforts emphasize improving material durability, optimizing thermal protection designs, and quantifying uncertainties in ablation modeling through Bayesian inference techniques.

Giorgos PetrouView profile

Research Fellow

• Building Physics
• Urban Modelling
• Indoor Environmental Quality
• +6 more

Giorgos Petrou is a Research Fellow in Building Physics and Urban Modelling at the UCL Institute for Environmental Design and Engineering, and a Doctoral Researcher (MPhil/PhD) at the UCL Energy Institute. His work focuses on integrating building simulation, environmental data, and health equity to address climate change impacts on indoor environments. He contributes to the NIHR-funded Health Protection Research Unit (HPRU) on Environmental Change and Health, examining indoor air quality and ventilation in residential settings. Giorgos holds a BSc in Physics from the University of Warwick, an MRes in Energy Demand Studies from UCL, and is nearing completion of a PhD on Bayesian calibration of archetype-based indoor temperature models. His research spans thermal simulation, data-driven statistical methods, and causal inference in built environment studies. He has presented at conferences including the AIVC Conference and Building Simulation and Optimisation Conference. His research interests emphasize the intersection of climate adaptation, housing policy, and health equity, with a focus on vulnerable populations such as care home residents. He explores trade-offs in natural ventilation strategies, model validation methodologies, and participatory approaches to policy design.

• Numerical Analysis
• Uncertainty Quantification
• Computational Mechanics
• +5 more

Prof. Raul Fidel Tempone is a renowned expert in Numerical Analysis and Uncertainty Quantification (UQ) at RWTH Aachen University, where he established the Lehrstuhl für Mathematics for Uncertainty Quantification . His research focuses on developing efficient numerical methods for stochastic models and differential equations, driven by applications in computational mechanics, quantitative finance, biological/chemical modeling, and wireless communication. Research emphasizes a posteriori error estimation, adaptive algorithms, Bayesian model calibration/validation, and optimal experimental design. Key contributions include multilevel Monte Carlo (MLMC) methods, hierarchical/sparse approximation, stochastic optimization, and machine learning integration for UQ. Recent work includes advancements in: MLMC for PDEs/SDEs and McKean-Vlasov equations Bayesian experimental design with nuisance parameters Uncertainty quantification in porous media and wireless networks Machine learning-based segmentation and filtering techniques He has advised numerous PhD/Master’s students and collaborates widely, producing over 150+ publications in top journals/conferences. His applied research bridges theoretical developments with real-world challenges in engineering, finance, and data science.

Jonathan OwenView profile

Academic

• Bayesian Statistics
• Uncertainty Quantification
• Climate Modeling
• +3 more

Jonathan Owen is a Research Associate in the School of Mathematical and Physical Sciences at the University of Sheffield. His current work focuses on the NERC-funded project 'Aerosol-MFR,' where he develops Bayesian statistical methods to reduce uncertainties in the Met Office's UK Earth System Model (UKESM1). He also collaborates with UN teams on epidemiological modeling and has prior experience with ice sheet and climate simulations through the UKRI Future Leaders Fellowship project. Jonathan holds a PhD in Mathematical Statistics from Durham University, supervised by Prof Ian Vernon and Prof Michael Goldstein, where he researched Bayesian uncertainty analysis for energy production optimization. He later worked on the JUNE model at Durham, applying Bayes linear emulation to study infectious disease transmission during the Covid-19 pandemic in England and refugee camps. His research interests span Bayesian emulation, Gaussian processes, uncertainty quantification, and model-observation comparison techniques. He specializes in applying these methods to environmental science challenges, including climatology, glaciology, and atmospheric sciences, with supplementary expertise in epidemiology and subsurface energy resource modeling. His publications highlight methodological innovations in Bayesian emulation for complex systems, such as handling partial discontinuities and hierarchical ensemble modeling. He emphasizes linking past, present, and future climate-ice sheet simulations to improve predictive accuracy for sea level rise and glacial dynamics. Notably, Jonathan collaborates with interdisciplinary teams across institutions like the University of Exeter and the Met Office, contributing to projects funded by NERC and UKRI. His work bridges statistical theory with real-world applications, particularly in decision support for environmental and energy systems.

Danfei XuView profile

Assistant Professor

• Robotics
• Artificial Intelligence
• Machine Learning
• +5 more

Dr. Danfei Xu is an Assistant Professor at the School of Interactive Computing at Georgia Tech and a part-time Research Scientist at NVIDIA AI. He holds a B.S. from Columbia University (2015) and a Ph.D. from Stanford University (2021), advised by Fei-Fei Li and Silvio Savarese. His research focuses on enabling robots to perform physical tasks in human environments through advancements in robotics, machine learning, and computer vision. Education: B.S. in Computer Science, Columbia University (2015) Ph.D. in Computer Science, Stanford University (2021) Research Interests: Robot Learning from Human Data (e.g., egocentric videos) Long-horizon reasoning via generative models Robot learning systems (hardware/software integration) Imitation learning, reinforcement learning, and visuomotor control Recent Articles Trends: His publications emphasize generative models for skill chaining, uncertainty-driven mapping, and human-in-the-loop learning. Key themes include leveraging diffusion models for complex manipulation tasks and developing open-source robotic systems. Awards and Grants: Meta Grant (2024) for robot learning from human data NSF Grant (2024) for generative models in robotics Autodesk Grant (2023) for high-precision manipulation IEEE RA-L Best Paper Honorable Mention (2024) ICRA 2024 Best Paper Award (Open-X-Embodiment) Advising and Teaching: Georgia Tech courses: Deep Learning (CS 7643/4644), Deep Learning for Robotics (CS 8803 DLM) Ph.D. recruitment in Robotics/ML/AI Lab: Robot Learning and Reasoning Lab (RL2) Labs and Teams: RL2 Lab develops adaptable robotic 'brains' for home, healthcare, and search/rescue Focus areas: End-to-end learning, generative models, and open-source systems

David OkayaView profile

Research Professor

• Structural Seismology
• Active-Source Seismology
• Crustal Physical Properties
• +5 more

David Okaya is an Associate Professor (Research) of Earth Sciences at the University of Southern California's Dornsife College of Letters, Arts and Sciences. His research focuses on structural seismology, subduction zone processes, and crustal anisotropy, with fieldwork in Taiwan, New Zealand, and the Pacific region. He has led major international projects like the TAIGER (Taiwan Arc-Collision Experiment) and SHIRE (Seismogenesis Hikurangi Integrated Research Experiment) initiatives, integrating seismic, geophysical, and geological data. Education: B.A. in Geology & Geophysics (Princeton University), M.S. and Ph.D. in Geophysics (Stanford University). Research Interests: Okaya’s work spans seismic imaging of plate boundaries, crustal anisotropy mechanisms, and thermal modeling of magmatic systems. Key areas include the Hikurangi Subduction Zone (New Zealand), Taiwan’s collision zone, and the Alpine Fault (South Island, New Zealand). He pioneered double-sided onshore-offshore seismic profiling techniques to map crustal structure across plate boundaries. Publications: Over 100 peer-reviewed articles, including seminal work on seismic anisotropy, subduction zone dynamics, and crustal deformation. Recent focus on the Hikurangi margin’s seismic structure and megathrust slip behavior. Awards: 1995 Fellow of the Geological Society of America. Fieldwork & Grants: Principal investigator for NSF-funded projects (e.g., SAHKE, SHIRE) and international collaborations with GNS Science (New Zealand), National Central University (Taiwan), and others. Extensive field experience in seismic experiments, including dynamite profiling in Taiwan and New Zealand. Labs/Teams: Core member of the Southern California Earthquake Center (SCEC) and contributor to the Community Modeling Environment, advancing physics-based seismic hazard analysis.

Xun HuanView profile

Associate Professor

• Uncertainty Quantification
• Data-Driven Modeling
• Numerical Optimization
• +3 more

Xun Huan is an Associate Professor in the Department of Mechanical Engineering at the University of Michigan, Ann Arbor. He holds a PhD in Computational Science and Engineering from MIT (2015), an SM in Aeronautics and Astronautics from MIT (2010), and a B.A.Sc. in Engineering Science (Aerospace) from the University of Toronto (2008). His research focuses on uncertainty quantification, data-driven modeling, numerical optimization, Bayesian analysis, and machine learning applications across engineering systems. Key research areas include optimal experimental design, stochastic modeling for manufacturing processes, and interdisciplinary applications in aerospace, energy systems, and biomedical engineering. He leads a research group conducting weekly meetings, emphasizing collaborative problem-solving and student independence. His work has been supported by grants such as the Keck Foundation award for cancer cell research (2022). He actively mentors PhD students, emphasizing continuous literature engagement, flexible work hours, and strategic internship planning. Education : Ph.D., MIT (2015) S.M., MIT (2010) B.A.Sc., University of Toronto (2008) Awards/Grants : Keck Foundation Collaborative Research Award (2022) Regents-approved promotion to Associate Professor (2024) Advising Philosophy : Weekly one-on-one meetings transitioning from hands-on to independent guidance Encourages academic career exploration through teaching opportunities Supports strategic internships, particularly in later PhD stages Current research emphasizes uncertainty-aware AI for digital twins, Bayesian methods in complex systems, and optimal experimental design frameworks. His group collaborates on projects ranging from material flow analysis to space weather prediction, leveraging interdisciplinary methods to address real-world challenges.

• Applied Mathematics
• Computational Science
• Plasma Physics
• +7 more

Giovanni Samaey is a Professor of Applied Mathematics and Mathematical Engineering at KU Leuven's Faculty of Engineering Science. He leads research in computational and multiscale methods, focusing on plasma edge modeling for nuclear fusion reactors, Bayesian inversion, and Monte Carlo algorithms. Appointed in 2011, he currently supervises ten PhD students and has held a five-year membership in the Young Academy. His work bridges academic research with societal impact, co-founding Platform Wiskunde Vlaanderen to strengthen mathematics in Flanders. Education: Graduated in Computer Science (specializing in applied mathematics) from KU Leuven (1996), completed a PhD in 2001 under Prof. Dirk Roose, supported by an NFWO fellowship. He transitioned from engineering studies due to a passion for mathematics' societal impact, initially avoiding academia but ultimately embracing teaching and research. Research Interests: Development of numerical methods for multiscale phenomena, including micro-macro acceleration algorithms, multilevel Monte Carlo techniques, and hybrid fluid-kinetic models. His work addresses challenges in plasma physics, fusion energy systems, and inverse problems. Key contributions include the X-Factor book (with Joos Vandewalle) promoting mathematics outreach and advancing computational tools for plasma edge simulations. Awards/Honors: Member of the Young Academy (2016-2021), NFWO Aspirant Fellowship (2001-2002). His efforts in STEM advocacy and mathematics promotion through Platform Wiskunde Vlaanderen highlight his dedication to education and public engagement. Advising & Leadership: Supervises a team of PhD students in applied mathematics and computational science. Active in curriculum development and interdisciplinary collaborations, particularly in fusion energy modeling. His research group contributes to codes like EMC3-EIRENE for plasma edge simulations. Labs/Teams: Leads projects in multiscale numerical methods and plasma simulation within KU Leuven's engineering faculty. Collaborates internationally on fusion reactor modeling and Monte Carlo algorithm design, emphasizing computational efficiency and scalability.

• Building Energy Efficiency
• HVAC Systems
• Heat Pump Technology
• +5 more

Jie Ma is a Research Fellow at the School of Mechanical Engineering, Purdue University, West Lafayette, specializing in advanced building energy systems and space habitat technologies through rigorous experimental and computational approaches. Research expertise includes: Building Energy Efficiency HVAC Systems Heat Pump Technology Thermal Comfort Compressor Technology Building Control Systems Cyber-Physical Systems Space Habitat Engineering Current work integrates cyber-physical testbeds for resilient space habitats with terrestrial building innovations, emphasizing field-validated solutions for energy efficiency and thermal performance. Publications demonstrate consistent focus on compressor degradation modeling, wall-integrated micro heat pumps, and advanced HVAC control strategies, reflecting strong industry collaboration and practical implementation focus.

Ross DavidsonView profile

Lecturer

• Simulation and mathematical models
• Epidemiology (livestock/wildlife/plant disease)
• Bayesian statistical inference
• +4 more

Ross Davidson is a Lecturer at Scotland's Rural College (SRUC), affiliated with the Animal and Veterinary Sciences School and Disease Systems department. He holds a PhD in Theoretical Physics from the University of Manchester (2003), focusing on stochastic systems and non-equilibrium statistical physics. His research bridges mathematical modeling with applied epidemiology, specializing in livestock and wildlife disease dynamics. Davidson collaborates widely across institutions including the University of York, Heriot-Watt University, and the Roslin Institute. He teaches multiple courses in veterinary epidemiology, statistics, and disease modeling at SRUC and the University of Edinburgh. His supervision record includes 10 PhD students focusing on topics like stochastic disease control, biosecurity modeling, and climate change impacts on parasites. Davidson's research interests center on epidemiological modeling, Bayesian inference applications, and livestock health management. He leads projects funded by Scottish Government and EU frameworks, such as socio-epidemiological approaches to livestock biosecurity and productive land management strategies. Key areas of investigation include anthelmintic efficacy, livestock trading networks, and environmental pathogen persistence. He has authored over 20 peer-reviewed articles since 2006, with recent work addressing mechanistic livestock disease models and the efficacy of European heather extracts against gastrointestinal nematodes. His contributions span from theoretical stochastic systems to applied agricultural epidemiology, emphasizing interdisciplinary approaches to disease control.

Larry WassermanView profile

Professor

• nonparametric inference
• causal inference
• machine learning
• +6 more

Larry Wasserman is a UPMC University Professor at Carnegie Mellon University, jointly appointed in the Department of Statistics and Data Science and the Machine Learning Department. He received his Ph.D. from the University of Toronto in 1988 and is recognized as one of the leading statisticians of his generation. His research spans theoretical and applied statistics, with core interests in: Foundational inference : Nonparametric methods, asymptotic theory, causal frameworks Modern applications : Machine learning, high-dimensional statistics, astrostatistics Interdisciplinary domains : Bioinformatics, genomics, physical sciences via the STAMPS group His recent publications demonstrate strong emphasis on causal methodology, optimal transport, and robust inference, with applications ranging from particle physics to genomic analysis. Articles frequently develop novel nonparametric techniques with minimax optimality guarantees. Award highlights include: COPSS Presidents' Award (1999) - Top honor for statisticians under 40 CRM-SSC Prize (2002) - Landmark contributions to statistics Fellowships: American Statistical Association, Institute of Mathematical Statistics, AAAS He leads the Statistical Machine Learning Theory Group and founded STAMPS (Statistical Methods for Physical Sciences). His textbooks All of Statistics and All of Nonparametric Statistics are widely used in graduate programs globally.

• Statistical Machine Learning
• High-Dimensional Statistics
• Nonparametric Methods
• +7 more

Ann Lee is a Professor and Co-Director of the Ph.D. Program in the Department of Statistics & Data Science at Carnegie Mellon University, with a joint appointment in the Machine Learning Department. She has been at CMU since 2005, after serving as the J.W. Gibbs Assistant Professor of Applied Mathematics at Yale University. Her academic journey began with a PhD in Physics from Brown University in 2002, preceded by BSc and MSc degrees in Engineering Physics from Chalmers University of Technology in Sweden. Dr. Lee's educational background includes: PhD in Physics, Brown University MSc/BSc in Engineering Physics, Chalmers University of Technology, Sweden Her research focuses on developing statistical methodology for complex data and problems in the physical sciences, with particular emphasis on trust-worthy scientific inference and reliable uncertainty quantification. She works at the intersection of classical statistics and machine learning, developing methods for simulation-based inference and experimental design. Dr. Lee is especially interested in likelihood-free inference, calibrated probabilistic forecasting, and interpretable diagnostics of generative models, with applications in astronomy and hurricane intensity guidance involving satellite imagery and large surveys. Dr. Lee co-founded the STAMPS (STAtistical Methods for the Physical Sciences) research group with Mikael Kuusela in 2018, which is transitioning to a CMU Research Center in Fall 2024. The group hosts public colloquia-style webinars and weekly research meetings for students and faculty at CMU and UPitt. Her notable scientific contributions have been recognized through: Her paper "Detecting Distributional Differences in Labeled Sequence Data with Application to Tropical Cyclone Satellite Imagery" being selected for The Best of AOAS session at the 2023 Joint Statistical Meeting Her student Luca Masserano winning the ASA Best Student Paper Award at the 2023 Joint Statistical Meeting Her student Alex Shen winning a Poster Award at the Machine Learning and the Physical Sciences Workshop, NeurIPS 2023 Dr. Lee has advised numerous PhD students to completion, with research spanning statistical machine learning, high-dimensional statistics, and applications in physical sciences. She has secured research funding supporting her work on statistical methods for physical sciences and has collaborated extensively with researchers across disciplines. Her current research projects focus on advancing likelihood-free inference methods and applying them to challenging problems in climate science and astronomy. She leads the STAMPS research group, which is becoming a CMU Research Center in Fall 2024, and organizes workshops including the PHYSTAT-SBI workshop on "Simulation Based Inference in Fundamental Physics" and the Hammers & Nails workshop on "Frontiers in Machine Learning in Cosmology, Astro & Particle Physics."

Emilio FerrerView profile

Professor

• Developmental Psychology
• Dynamical Systems
• Latent Growth Analysis
• +4 more

Emilio Ferrer is a Professor of Psychology at the University of California, Davis. His research focuses on developmental methodology, emphasizing techniques to analyze change and intra-individual variability, such as latent growth analysis and both linear and nonlinear dynamical systems. His work frequently explores the dynamics of affective processes in dyadic interactions, with applications in cognitive neuroscience, gerontology, and clinical psychology. Recent publications highlight his contributions to modeling nonlinear developmental processes across diverse contexts, including autism symptom trajectories, meditation's long-term effects, and physiological linkages in caregiving dyads. Ferrer’s methodological innovations include state space models, Bayesian hierarchical frameworks, and computational tools for simulating psychological theories. His research spans interdisciplinary domains, integrating neuroscience, behavioral science, and public health. Collaborations with experts in psychiatry, neurology, and biostatistics underscore his commitment to advancing longitudinal data analysis techniques. Despite the absence of explicitly listed awards or students, his prolific publication record reflects significant influence in developmental and computational psychology.