MIT Uncertainty Quantification Group

Youssef Marzouk is a Professor in the Department of Aeronautics and Astronautics at MIT, and co-director of the MIT Center for Computational Science & Engineering. He is also a core member of MIT's Statistics and Data Science Center and director of the MIT Aerospace Computational Design Laboratory. His research interests lie at the intersection of computation and statistical inference with physical modeling. He develops new methodologies for uncertainty quantification, Bayesian modeling and computation, data assimilation, experimental design, and machine learning in complex physical systems. His methodological work is motivated by a wide variety of engineering, environmental, and geophysics applications. He received his SB, SM, and PhD degrees from MIT and spent four years at Sandia National Laboratories before joining the MIT faculty in 2009. He is a recipient of the Hertz Foundation Doctoral Thesis Prize, the Sandia Laboratories Truman Fellowship, the US Department of Energy Early Career Research Award, and the Junior Bose Award for Teaching Excellence from the MIT School of Engineering. He is an Associate Fellow of the AIAA and currently serves on the editorial boards of the SIAM Journal on Scientific Computing, the SIAM/ASA Journal on Uncertainty Quantification, and several other journals. He is also an avid coffee drinker and occasional classical pianist.

I am interested in problems that lie at the intersection of applied mathematics, signal processing, and machine learning. Currently, I am researching new representations of sets and probability measures, with a focus on their application in the field of optimal experimental design. I completed my Ph.D. in theoretical signal processing and machine learning at Centrale Lille, under Pierre Chainais and Rémi Bardenet. My work includes subset selection problems, quadrature rules and interpolation methods using determinantal point processes. Before joining MIT, I collaborated with Rémi Gribonval on compressive learning using sketching methods at ENS de Lyon. In my free time, I enjoy reading, traveling, and playing tennis.

My research is broadly in the area of numerical methods for uncertainty quantification and mathematical modeling. I am currently interested in techniques that exploit multi-scale/multi-feature structure in a system's dynamics. These methods accelerate large scale computations. Before joining MIT, I received an M.S. in Mathematics from Virginia Tech, where I studied nonlinear eigenvalue problems and reduced order modeling. Outside of work, I enjoy cooking, snowboarding, and exploring Boston.

Jan Glaubitz is a Postdoctoral Associate in the Department of Aeronautics and Astronautics at MIT, collaborating with Professor Youssef Marzouk. His research aims to advance the foundational computational methodologies in numerical conservation laws and hierarchical Bayesian learning. Passionate about the interplay between theoretical numerical analysis, method development, and uncertainty quantification, he strives to establish provable approximation, convergence, and stability results while quantifying confidence in computational predictions.

I am broadly interested in Bayesian Inference, Data Assimilation, and Fluid Mechanics. I have a particular interest in exploiting structures of forward and inverse problems to develop robust and scalable methods. Before joining MIT, I completed my PhD in Mechanical Engineering at UCLA, advised by Jeff Eldredge. Outside of work, I enjoy arts, outdoor activities, and cooking.

I am a Postdoctoral Fellow at the Broad Institute of MIT/Harvard in the Eric and Wendy Schmidt Center. I am broadly interested in intersections of machine learning, dynamical systems, and biomedical sciences. My work focuses on improving the prediction and inference of biological and physical systems by blending machine learning, mechanistic modeling, and data assimilation techniques. I studied biophysics as an undergraduate at Columbia University, and did a PhD in Computing and Mathematical Sciences at Caltech under the supervsion of Andrew Stuart. Outside of work, I enjoy playing music, going to concerts, playing tennis, skiing, and camping.

I am a computational scientist/applied mathematician with interests in scientific inverse problems. Before working with Youssef I completed a PhD in computational science and engineering at MIT, advised by Laurent Demanet. Prior to MIT, I attended the University of Toronto and graduated with a bachelors and a masters degree in aerospace engineering. I enjoy watching and playing hockey, and it has been a painful experience being a fan of the Toronto Maple Leafs while living in Boston all these years.

My research interests lie in the field of scientific computing, uncertainty quantification and machine learning in applications of environmental fluids. I am working with Youssef and Ruben Juanes (MIT CEE) on leveraging advances in data-driven tools to improve the efficiency of computation, quantify the uncertainty of models, and provide scientific information to inform decision-making on climate change. Before joining MIT, I completed my PhD in energy resources engineering at Stanford, advised by Daniel M. Tartakovsky. Outside of work, I enjoy playing the harp, travelling and exploring good restaurants.

My research interests are on high dimensional (tensor) data analysis, regularization for inverse problems, uncertainty quantification, and high performance computing. One of my main research goals is in developing computationally efficient numerical methods to solve large-scale inverse problems. Such problems arise from an extended list of applications in data science and engineering, where from limited observations collected from surface measurements and knowledge of the forward process that maps the unknown parameters onto the data, the goal is to reconstruct the unknown parameters. Those methods commonly rely on numerical linear algebra, but I also use techniques and tools from statistics, numerical optimization, machine learning, and PDEs.

My current research here at MIT focuses on entropic optimal transport methods for conditional simulations in the context of Bayesian inverse problems. Previously, during my PhD at Ruhr University Bochum, I have worked on extreme event quantification for stochastic (partial) differential equations using large deviation theory. Outside of work, I enjoy reading, watching movies and playing badminton.

My research interests focus on the development of a codebase for atomistic simulations as part of the CESMIX project. This will include developing a toolkit of various active learning schemes to fit machine-learned interatomic potentials, as well as facilitating forward uncertainty propagation across multiscale simulations of complex materials processes. Prior to joining the group, I completed my PhD in materials science at MIT under Asegun Henry; my thesis work focused on improving molecular dynamics simulations of interfacial heat transport. Outside of work, I enjoy the standard affair of watching movies (especially on Criterion), reading, visiting art museums, and eating at restaurants.

My research interests include optimal experimental design in the presence of model error and other topics in uncertainty quantification. I received my bachelor's degree in Physics from the California Institute of Technology. Outside of my work, I am interested in classical piano, photography, web design, and the culinary arts.

My current research is focused on uncertainty quantification for density functional theory. Broadly, my interests include Bayesian inference, multifidelity modeling, and uncertainty propagation over multiple scales. I graduated from the University of Texas at Austin with bachelor’s degrees in computational engineering and mathematics. I completed an undergraduate thesis related to feminist media studies. In my free time, I enjoy fiber arts, reading, and alpine slides.

I am broadly interested in applied probability, machine learning, and optimization. My current research focuses on efficient rare event simulation and constrained transport-based density estimation. I have worked in randomized numerical linear algebra (CMU), scalable experiment design (Google), and disease classification (Petuum). Before joining MIT, I received a bachelors in EECS from the National Technical University of Athens, as well as a masters in machine learning and a masters in computer science from CMU. I also enjoy hiking, photography, soccer and exploring new restaurants.

My research is focused on Bayesian inference and uncertainty quantification for high-dimensional problems. I’m currently working on developing algorithms in Earth remote sensing for climate science. I grew up in Canada and graduated from the Engineering Science program at the University of Toronto in 2019, majoring in aerospace engineering. Outside of work, I enjoy travel and culinary adventures.

My current research is related to generative modeling and optimal experimental design. In addition, I am interested in Bayesian statistics, optimization and applied probability. I was born and raised in Zhengzhou, China and lived in College Station, Texas for a few years, where I received my Bachelor’s degrees in Mathematics and Mechanical Engineering from Texas A&M University. In my free time, I enjoy playing tennis and listening to music.

My research interests are in mathematical and physical modeling, computational techniques and applied probability. My current research focuses on the long-time simulation of molecular systems for the CESMIX project, and involves stochastic modeling, high-dimensional approximation and numerical integration. Before joining MIT, I grew up in Italy and France, and graduated in Mathematics and Physics, at Ecole Polytechnique, Paris. In my free time, I enjoy playing soccer, tasting French wine and cooking Italian food!

My current research is focused on developing iterative methods for Bayesian inference using ideas from geometry and dynamics. I additionally have interests in data assimilation, covariance estimation, and application of Bayesian inference to physical systems. Prior to joining the UQ Group I spent 1.5 years on the technical staff at MIT Lincoln Laboratory, and I have a B.S. in Mathematics and a B.S. in Computational Modeling and Data Analytics from Virginia Tech (2019). In my free time I can often be found swimming, cooking, cycling, reading, or crafting.

My research interest broadly lies in mathematical problems arising from data science. In particular, I am interested in using tools from variational analysis, empirical process theory, and PDEs to develop the mathematical theory of learning and explore its implications for possible advancement in numerical algorithms. My current research projects involve quantifying approximation rates for ODE-parametrized transport maps and establishing sample complexity bounds for sampling using NeuralODEs.

I'm currently researching online parameter estimation using data assimilation in the context of wind fields, with prior experience in high performance computing, model order reduction, and other topics. I'm broadly interested in making simulations both fast and robust using surrogates, code optimizations, software tools, and theory. I grew up near Raleigh, NC, but went to Virginia Tech for a B.S. in Computational Modeling and Data Analytics. For fun, I enjoy perfecting the art of making/eating food, watching films, and meticulously digging in stacks of records.

I am interested in problems that lie in the intersection of stochastic analysis, differential geometry and inference with applications to uncertainty quantification and physics. I am currently researching optimal transport via the lens of information geometry. Before joining the UQ group I was an undergraduate student in Columbia University where I double majored in Mathematics and Physics. In my free time I enjoy exercising, being outdoors and visiting art exhibits.

My research interest is in the combination of statistical inference methods and first-principles physical modeling to improve prediction under uncertainty. Currently, I work on applying Bayesian inference techniques to atomistic modeling of materials in the CESMIX project. From my previous positions as a research fellow at TU Delft and MS student at Stanford, I have experience in data assimilation, surrogate modeling, and probabilistic risk analysis. Outside of work, I enjoy rock climbing, hiking, tennis, and live music.

My current research revolves around developing algorithms that enable accurate, efficient, and systematic sampling from conditional distributions arising from complex dynamical systems with physical constraints. Beyond developing such algorithms, I am researching conditional robustness for inference and sampling methods through establishing guarantees under modeling uncertainty. Prior to joining MIT, I pursued my undergraduate studies at Emory University, majoring in applied mathematics and statistics (AMS). In my free time, I enjoy alpine skiing, golf, and food explorations.

My research interest is three-fold: applied probability, Bayesian inverse problems, and scientific computing. I am excited about everything with a random nature, and developing methods to explore the universal governing rules beneath the chaos. Before joining MIT, I graduated from New York University Shanghai with bachelor’s degrees in Mathematics and Data Science. Besides research, I am a big fan of reading, outdoor activities, classical music, and making food maps around Boston.