BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Michigan Institute for Computational Discovery and Engineering - ECPv6.17.1//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-ORIGINAL-URL:https://micde.umich.edu
X-WR-CALDESC:Events for Michigan Institute for Computational Discovery and Engineering
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/Detroit
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20230312T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20231105T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20240310T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20241103T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20250309T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20251102T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20260308T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20261101T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20250711T120000
DTEND;TZID=America/Detroit:20250711T130000
DTSTAMP:20250808T204538Z
CREATED:20250708T154831Z
LAST-MODIFIED:20250808T204538Z
UID:10000826-1752235200-1752238800@micde.umich.edu
SUMMARY:Frontiers in Scientific Machine Learning Seminar 17: SALSA-RL: Stability Analysis in the Latent Space of Actions for Reinforcement Learning
DESCRIPTION:Date: June 6\, 2025\, 12pm – 1pm\nThis is a hybrid event. To join via Zoom: Meeting ID: 978 2352 7756\, Passcode: Enter last year in format YYYY\nTo join in person: 2004 Lay Auto Lab. Refreshments will be available! \nAbstract: Modern deep reinforcement learning (DRL) methods have made significant advances in handling continuous action spaces. However\, real-world control systems–especially those requiring precise and reliable performance–often demand interpretability in the sense of a-priori assessments of agent behavior to identify safe or failure-prone interactions with environments. To address this limitation\, we propose SALSA-RL (Stability Analysis in the Latent Space of Actions)\, a novel RL framework that models control actions as dynamic\, time-dependent variables evolving within a latent space. By employing a pre-trained encoder-decoder and a state-dependent linear system\, our approach enables interpretability through local stability analysis\, where instantaneous growth in action-norms can be predicted before their execution. We demonstrate that SALSA-RL can be deployed in a non-invasive manner for assessing the local stability of actions from pretrained RL agents without compromising on performance across diverse benchmark environments. By enabling a more interpretable analysis of action generation\, SALSA-RL provides a powerful tool for advancing the design\, analysis\, and theoretical understanding of RL systems.\nBio: Romit Maulik is an Assistant Professor in the College of Information Sciences and Technology at Pennsylvania State University (Penn State). He is also a co-hire in the Institute for Computational and Data Sciences at Penn State and a Joint Appointment Faculty at Argonne National Laboratory. He obtained his PhD in Mechanical and Aerospace Engineering at Oklahoma State University (in 2019) and was the Margaret Butler Postdoctoral Fellow (from 2019-2021) before becoming an Assistant Computational Scientist at Argonne National Laboratory (from 2021-2023). His group studies high-performance multifidelity scientific machine learning algorithm development with applications to various multiphysical nonlinear dynamical systems such as those that arise in fluid dynamics\, geophysical modeling\, nuclear fusion\, and beyond. He is an Early Career Awardee of the Army Research Office.
URL:https://micde.umich.edu/event/workshop-seminarfrontiers-in-scientific-machine-learning-seminar-17-salsa-rl-stability-analysis-in-the-latent-space-of-actions-for-reinforcement-learning/
LOCATION:Walter E Lay Auto Lab – 2004
CATEGORIES:Deep Learning,FSML,Interdisciplinary,Machine Learning,North Campus,Scientific Computing,Sciml
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20250711T120000
DTEND;TZID=America/Detroit:20250711T130000
DTSTAMP:20250708T153951Z
CREATED:20250708T153951Z
LAST-MODIFIED:20250708T153951Z
UID:10000825-1752235200-1752238800@micde.umich.edu
SUMMARY:Frontiers in Scientific Machine Learning Seminar - Romit Maulik (Penn State University): SALSA-RL: Stability Analysis in the Latent Space of Actions for Reinforcement Learning
DESCRIPTION:Zoom link \nBio: Romit Maulik is an Assistant Professor in the College of Information Sciences and Technology at Pennsylvania State University (Penn State). He is also a co-hire in the Institute for Computational and Data Sciences at Penn State and a Joint Appointment Faculty at Argonne National Laboratory. He obtained his PhD in Mechanical and Aerospace Engineering at Oklahoma State University (in 2019) and was the Margaret Butler Postdoctoral Fellow (from 2019-2021) before becoming an Assistant Computational Scientist at Argonne National Laboratory (from 2021-2023). His group studies high-performance multifidelity scientific machine learning algorithm development with applications to various multiphysical nonlinear dynamical systems such as those that arise in fluid dynamics\, geophysical modeling\, nuclear fusion\, and beyond. He is an Early Career Awardee of the Army Research Office. \nSALSA-RL: Stability Analysis in the Latent Space of Actions for Reinforcement Learning\nAbstract: Modern deep reinforcement learning (DRL) methods have made significant advances in handling continuous action spaces. However\, real-world control systems–especially those requiring precise and reliable performance–often demand interpretability in the sense of a-priori assessments of agent behavior to identify safe or failure-prone interactions with environments. To address this limitation\, we propose SALSA-RL (Stability Analysis in the Latent Space of Actions)\, a novel RL framework that models control actions as dynamic\, time-dependent variables evolving within a latent space. By employing a pre-trained encoder-decoder and a state-dependent linear system\, our approach enables interpretability through local stability analysis\, where instantaneous growth in action-norms can be predicted before their execution. We demonstrate that SALSA-RL can be deployed in a non-invasive manner for assessing the local stability of actions from pretrained RL agents without compromising on performance across diverse benchmark environments. By enabling a more interpretable analysis of action generation\, SALSA-RL provides a powerful tool for advancing the design\, analysis\, and theoretical understanding of RL systems.
URL:https://micde.umich.edu/event/frontiers-in-scientific-machine-learning-seminar-romit-maulik-penn-state-university/
LOCATION:2004 Lay Auto Lab
CATEGORIES:Ai In Science And Engineering,Computational Modeling,Engineering,FSML,Graduate School,Interdisciplinary,North Campus,Research,Sciml,Statistics
ATTACH;FMTTYPE=image/png:https://micde.umich.edu/wp-content/uploads/2025/07/Romit-Maulik-PennState-University.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20250620T120000
DTEND;TZID=America/Detroit:20250620T130000
DTSTAMP:20250620T145718Z
CREATED:20250619T132221Z
LAST-MODIFIED:20250620T145718Z
UID:10000823-1750420800-1750424400@micde.umich.edu
SUMMARY:Frontiers in Scientific Machine Learning Seminar - Pan Du (University of Notre Dame): Conditional neural field latent diffusion model for generating spatiotemporal turbulence
DESCRIPTION:Zoom link \nBio: Pan Du received his bachelor’s degree in Thermal Engineering from Tsinghua University and completed his master’s in Mechanical Engineering at Washington University in St. Louis. He is currently a Ph.D. candidate in Aerospace and Mechanical Engineering at the University of Notre Dame under the guidance of Prof. Jian-Xun Wang. Pan’s research spans multiple disciplines\, including scientific machine learning\, Bayesian inference\, uncertainty quantification\, geometric deep learning\, and computational fluid mechanics. \nConditional neural field latent diffusion model for generating spatiotemporal turbulence\nAbstract: Pan Du will present the CoNFiLD model\, a novel generative framework for simulating complex turbulent flows in 3D irregular domains. While traditional eddy-resolved simulations are accurate\, their high computational cost limits usability. CoNFiLD addresses this by integrating neural field encoding with latent diffusion\, enabling efficient\, probabilistic modeling of spatiotemporal dynamics. It supports a wide range of tasks—such as flow super-resolution\, sparse reconstruction\, and data restoration—via Bayesian conditional sampling\, all without retraining. Results across diverse turbulent scenarios highlight its potential for advancing data-driven turbulence modeling.
URL:https://micde.umich.edu/event/frontiers-in-scientific-machine-learning-seminar-pan-du-university-of-notre-dame/
LOCATION:GG Brown Laboratory – 1642
CATEGORIES:Ai In Science And Engineering,Computational Modeling,Engineering,FSML,Graduate School,Interdisciplinary,North Campus,Research,Sciml,Statistics
ATTACH;FMTTYPE=image/png:https://micde.umich.edu/wp-content/uploads/2025/06/Pan-Du-1.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20250606T120000
DTEND;TZID=America/Detroit:20250606T130000
DTSTAMP:20250604T181957Z
CREATED:20250602T103951Z
LAST-MODIFIED:20250604T181957Z
UID:10000822-1749211200-1749214800@micde.umich.edu
SUMMARY:Frontiers in Scientific Machine Learning Seminar - Ashwin Renganathan (Penn State): Sample-efficient and Principled Decision-making  with Expensive Stochastic Oracles
DESCRIPTION:Zoom link \nBio: Ashwin Renganathan is an assistant professor of aerospace engineering at Penn State and holds a joint appointment with the Penn State Institute of Computational and Data Sciences (ICDS). He directs the Computational complex engineered Systems Design Laboratory (CSDL) at Penn State. He is broadly interested in developing novel and scalable computational techniques for surrogate modeling\, uncertainty quantification\, and numerical optimization\, with a focus on aerospace applications. He earned his Ph.D. in aerospace engineering from Georgia Tech and previously completed a postdoctoral appointment in applied mathematics at the Argonne National Laboratory. \nSample-efficient and Principled Decision-making with Expensive Stochastic Oracles\nAbstract: Modern day engineering decision-making involves one or more computer simulation oracles of an engineered system which can be queried on-demand to learn the system response to control input. Querying simulation oracles\, also called “computer experiments”\, incur a non-trivial computational cost\, which increases with the level of fidelity in the underlying models. For instance\, a realistic computational aerodynamic simulation of an aircraft can cost several thousands of CPU hours to compute—anything more than a few dozens of such simulations is prohibitive. Therefore\, a central goal of engineering decision-making is to optimally design computer experiments\, to maximize the value of information extracted at minimal computational effort.\nIn this talk\, we will address problems anchored in\, what we coin\, the “decision-making triad” which includes: surrogate modeling\, uncertainty quantification (UQ)\, and numerical optimization/control. Specifically\, using variants of a probabilistic surrogate model and a Bayesian decision theoretic framework\, we will show that problems in the decision-making triad can be solved in a principled\, theoretically sound and\, yet (computational) cost-effective manner. We will show demonstrations on applications in computational aerodynamics.
URL:https://micde.umich.edu/event/workshop-seminarfrontiers-in-scientific-machine-learning-seminar-15-ashwin-renganathan/
LOCATION:GG Brown Laboratory – 1642
CATEGORIES:Ai In Science And Engineering,Computational Modeling,Engineering,FSML,Graduate School,Interdisciplinary,North Campus,Research,Sciml,Statistics
ATTACH;FMTTYPE=image/png:https://micde.umich.edu/wp-content/uploads/2025/06/MICDE-Seminar-Series-Speaker-Portraits-4.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20250523T120000
DTEND;TZID=America/Detroit:20250523T130000
DTSTAMP:20250522T231313Z
CREATED:20250522T231313Z
LAST-MODIFIED:20250522T231313Z
UID:10000821-1748001600-1748005200@micde.umich.edu
SUMMARY:FSML Lecture Series - Smita Krishnaswamy (Yale University): Dynamics Models of Cellular and Neuronal Interactions
DESCRIPTION:Zoom link \nBio: Smita Krishnaswamy is an Associate Professor in the departments of Computer Science (SEAS) and Genetics (YSM). She is part of the programs in Applied Mathematics\, Computational Biology & Bioinformatics and Interdisciplinary Neuroscience. She is also affiliated with the Yale Institute for the foundations of data science\, Wu-Tsai Institute\, Yale Cancer Center. Her lab works on fundamental deep learning and machine learning developments for representing and learning from big data. Her techniques incorporate mathematical priors from graph spectral theory\, manifold learning\, signal processing\, and topology into machine learning and deep learning frameworks\, in order to denoise and model the underlying systems faithfully for predictive insight. Currently her methods are being widely used for data denoising\, visualization\, generative modeling\, dynamics. modeling\, comparative analysis and domain transfer in datasets arising from stem cell biology\, cancer\, immunology and structural biology (among others). \nPrior to joining Yale\, she completed her postdoctoral training at Columbia University in the systems biology department where she focused on learning computational models of cellular signaling from single-cell mass cytometry data. She obtained her Ph.D. from EECS department at University of Michigan where her research focused on algorithms for automated synthesis and probabilistic verification of nanoscale logic circuits. Following her time in Michigan\, she spent 2 years at IBM’s TJ Watson Research Center as a researcher in the systems division where she worked on automated bug finding and error correction in logic. Her work over the years has won several awards including the NSF CAREER Award\, Sloan Faculty Fellowship\, and Blavatnik fund for Innovation. \nDynamics Models of Cellular and Neuronal Interactions
URL:https://micde.umich.edu/event/fsml-lecture-series-smita-krishnaswamy/
LOCATION:GG Brown Laboratory – 1642
CATEGORIES:Ai In Science And Engineering,Artificial Intelligence,big data,College Of Engineering,data,FSML,Machine Learning,North Campus,Statistics
ATTACH;FMTTYPE=image/png:https://micde.umich.edu/wp-content/uploads/2025/05/MICDE-Seminar-Series-Speaker-Portraits-3.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20250425T120000
DTEND;TZID=America/Detroit:20250425T130000
DTSTAMP:20250522T231416Z
CREATED:20250421T155355Z
LAST-MODIFIED:20250522T231416Z
UID:10000819-1745582400-1745586000@micde.umich.edu
SUMMARY:FSML Lecture Series - Julie Bessac (National Renewable Energy Laboratory): Statistical learning for Summary Statistics of Physics-based Model Outputs and their Correction and Probabilistic Outputs from Neural Networks applied to Super-resolution
DESCRIPTION:Zoom link \nBio: Julie Bessac received her Ph.D. degree in 2014 in Applied Mathematics from the University of Rennes 1\, France. Between 2014 and 2023\, she was a post-doctoral appointee and a research scientist in the Mathematics and Computer Science Division at Argonne National Laboratory. She joined National Renewable Energy Laboratory in 2023 as a computational statistician. She is an adjunct professor at the Department of Statistics at Virginia Tech. Her research focuses on statistical and machine learning methods for modeling\, forecasting and uncertainty quantification for diverse applications: geophysical processes and their applications to energy systems\, computer science and nuclear physics. \nSummary Statistics of Physics-based Model Outputs and their Correction and Probabilistic Outputs from Neural Networks applied to Super-resolution\nAbstract: In the first part of this talk\, we will discuss summary statistics of physics-based model outputs and their correction with observational data. Physics-based models capture broad-scale dynamics across various spatial and temporal scales\, they often face challenges such as modeling biases\, high computational costs\, along with large outputs that are challenging to manipulate. On the other hand\, observations capture localized variability but are typically sparse. This talk presents an innovative approach to address these challenges by utilizing summary statistics from physics-based model outputs and enhancing them with observational information via neural networks.\nIn the second part of the talk\, we will present neural networks with closed-form probabilistic loss that applied to super-resolution of surface wind speed. We will illustrate that the use of a closed-form probabilistic loss provides the neural network with a sampling capability and a spatial covariance for super-resolved wind fields.\nThese are joint work with Atlanta Chakraborty (NREL)\, Harrison Goldwyn (NREL)\, Daniel Getter (USC)\, Johann Rudi (Virginia Tech) and Mitchell Krock (University of Missouri).
URL:https://micde.umich.edu/event/fsml-lecture-13-julie-bessac/
LOCATION:GG Brown Laboratory – 2636
CATEGORIES:Ai In Science And Engineering,Artificial Intelligence,big data,College Of Engineering,data,FSML,Machine Learning,North Campus,Statistics
ATTACH;FMTTYPE=image/png:https://micde.umich.edu/wp-content/uploads/2025/04/MICDE-Seminar-Series-Speaker-Portraits-1.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20250328T120000
DTEND;TZID=America/Detroit:20250328T130000
DTSTAMP:20250324T152136Z
CREATED:20250324T152136Z
LAST-MODIFIED:20250324T152136Z
UID:10000816-1743163200-1743166800@micde.umich.edu
SUMMARY:FSML Lecture Series - Lianghao Cao (Caltech): Derivative-Informed Operator Learning with Applications to Cost-Efficient Bayesian Inversion
DESCRIPTION:Zoom link \nBio: Dr. Lianghao Cao is a Postdoctoral Scholar Research Associate from the Department of Computing and Mathematical Sciences at the California Institute of Technology. He obtained a B.S. in Engineering Mechanics from the University of Illinois at Urbana-Champaign and a Ph.D. in Computational Science\, Engineering\, and Mathematics from The University of Texas at Austin. His research blends mechanistic modeling\, uncertainty quantification\, and scientific machine learning to understand\, enhance\, and control the quality\, validity\, and reliability of simulation-based predictions of complex physical systems. \nDerivative-Informed Operator Learning with Applications to Cost-Efficient Bayesian Inversion\nAbstract: This talk focuses on a derivative-informed supervised learning method for efficiently building machine learning surrogates of high-fidelity computational models\, particularly those governed by parametric partial differential equations. Unlike the conventional supervised learning method that treats the model as a black box\, our approach leverages additional model sensitivity information\, extracted via solving forward or adjoint sensitivity equations. This sensitivity information is integrated into the surrogate’s architecture and training process based on rigorous error analysis. We refer to such a surrogate construction as DINO (derivative-informed neural operator). \nDINO offers two key advantages over conventional surrogate construction. First\, it significantly improves the cost-accuracy trade-off for a wide range of models\, often by one to two orders of magnitude. Second\, it directly controls the surrogate Jacobian (Fréchet derivative) errors\, thus enhancing performance in surrogate-driven outer-loop problems that use gradient- and Hessian-based optimization algorithms. We demonstrate DINO’s capability to accelerate infinite-dimensional Bayesian inversion. First\, we show that geometric MCMC driven by DINO achieves a 2–9x speed up in asymptotically exact posterior sampling. Second\, we introduce LazyDINO\, a DINO-driven measure transport method for amortized Bayesian inversion\, which is one to two orders of magnitude more cost-efficient than competing methods.\nThis talk is based on joint work with Michael Brennan\, Joshua Chen\, Omar Ghattas\, Youssef Marzouk\, and Thomas O’Leary-Roseberry.
URL:https://micde.umich.edu/event/fsml-lecture-series-lianghao-cao-caltech-derivative-informed-operator-learning-with-applications-to-cost-efficient-bayesian-inversion/
LOCATION:2636 GGBA\, 2350 Hayward St\, Ann Arbor\, MI\, United States
CATEGORIES:Engineering,FSML,Science
ATTACH;FMTTYPE=image/png:https://micde.umich.edu/wp-content/uploads/2025/03/Lianghao-Cao-Caltech.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20250314T120000
DTEND;TZID=America/Detroit:20250314T130000
DTSTAMP:20250311T130831Z
CREATED:20250311T130516Z
LAST-MODIFIED:20250311T130831Z
UID:10000814-1741953600-1741957200@micde.umich.edu
SUMMARY:FSML Lecture Series - Alexander Tong (Mila - Quebec AI Institute): Flow matching in cell trajectories and protein design
DESCRIPTION:Zoom link \nBio: Alexander Tong is a postdoctoral fellow at Mila with Yoshua Bengio\, visiting researcher at Oxford with Michael Bronstein\, cofounder of Dreamfold—a protein design startup\, and incoming assistant professor at Duke University starting July 2025. Alex completed his Ph.D. in Computer Science at Yale University in 2021 with Smita Krishnaswamy. His research interests span generative modeling\, graph signal processing\, and optimal transport to understand biological systems with a focus on cells and proteins. \nFlow matching in cell trajectories and protein design\nAbstract: Generative flow models learn a (possibly stochastic) mapping between source and target distributions. Common paradigms include diffusion models\, score matching models\, and continuous normalizing flows. In this talk I will first present methods for improved training of flow models using flow matching objectives using ideas from optimal transport. I will then show how these improved methods can be applied to the tasks of (1) modelling cell dynamics\, which allow us to better understand disease programs –leading to a new potential therapeutic pathway for triple-negative breast cancer and (2) generative protein design\, with applications to biologic drug discovery.
URL:https://micde.umich.edu/event/fsml-lecture-series-alexander-tong-mila/
LOCATION:1642 GGBL\, 2350 HAYWARD ST\, Ann Arbor\, 48109\, United States
CATEGORIES:Engineering,FSML,Science
ATTACH;FMTTYPE=image/png:https://micde.umich.edu/wp-content/uploads/2025/03/Alexander-Tong.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20250228T120000
DTEND;TZID=America/Detroit:20250228T130000
DTSTAMP:20250227T225237Z
CREATED:20250225T171723Z
LAST-MODIFIED:20250227T225237Z
UID:10000809-1740744000-1740747600@micde.umich.edu
SUMMARY:FSML Lecture Series - Doruk Aksoy: From Turbulent Flows to Video Games: Managing Large-Scale Data with Tensor Decomposition
DESCRIPTION:Zoom link \nBio:Doruk Aksoy is a 5th year PhD candidate in Aerospace Engineering and Scientific computing at the University of Michigan\, working under the supervision of Prof. Alex Gorodetsky. Prior to joining UM\, he studied Mechanical Engineering at Bogazici University in Istanbul Turkey. During his PhD\, he worked on developing incremental tensor decomposition algorithms to accelerate scientific machine learning through data reduction. \nFrom Turbulent Flows to Video Games: Managing Large-Scale Data with Tensor Decomposition\nAbstract: The rapid advancement of large-scale parallel computing created a surge of interest in developing high-fidelity digital twins for complex systems. However\, the computational demands for training these models are immense\, requiring vast amounts of data. As the spatial and temporal resolution of simulations increases\, even data storage becomes a critical bottleneck. This talk presents how low-rank tensor decomposition methods can be used to exploit the structure in large-scale data. We showcase a diverse array of applications\, from 3D turbulent Navier-Stokes simulations to Minecraft gameplay videos\, demonstrating the versatility and power of these techniques.
URL:https://micde.umich.edu/event/fsml-lecture-series-doruk-aksoy/
LOCATION:2636 GGBA\, 2350 Hayward St\, Ann Arbor\, MI\, United States
CATEGORIES:Engineering,FSML,Science
ATTACH;FMTTYPE=image/png:https://micde.umich.edu/wp-content/uploads/2023/09/Aksoy-Doruk.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20250214T093000
DTEND;TZID=America/Detroit:20250214T103000
DTSTAMP:20250212T173805Z
CREATED:20250212T173700Z
LAST-MODIFIED:20250212T173805Z
UID:10000808-1739525400-1739529000@micde.umich.edu
SUMMARY:FSML Lecture Series - Ricardo Vinuesa: Identifying coherent structures and controlling turbulent flows through deep learning
DESCRIPTION:Zoom link \nBio: Dr. Ricardo Vinuesa is joining the Department of Aerospace Engineering at the University of Michigan in the Fall of 2025. He is currently an Associate Professor at the Department of Engineering Mechanics\, KTH Royal Institute of Technology in Stockholm. He studied Mechanical Engineering at the Polytechnic University of Valencia (Spain)\, and he received his PhD in Mechanical and Aerospace Engineering from the Illinois Institute of Technology in Chicago. His research combines numerical simulations and data-driven methods to understand\, control and predict complex wall-bounded turbulent flows\, such as the boundary layers developing around wings and urban environments. Dr. Vinuesa has received\, among others\, an ERC Consolidator Grant\, the TSFP Kasagi Award\, the MST Emerging Leaders Award\, the Goran Gustafsson Award for Young Researchers\, the IIT Outstanding Young Alumnus Award\, the SARES Young Researcher Award and he leads several large Horizon Europe projects. He is also a member of the Young Academy of Science of Spain. \nIdentifying coherent structures and controlling turbulent flows through deep learning\nAbstract: In this work we first use explainable deep learning based on Shapley explanations to identify the most important regions for predicting the future states of a turbulent channel flow. The explainability framework (based on gradient SHAP) is applied to each grid point in the domain\, and through percolation analysis we identify coherent flow regions of high importance. These regions have around 70% overlap with the intense Reynolds-stress (Q) events in two-dimensional vertical planes. Interestingly\, these importance-based structures have high overlap with classical turbulence structures (Q events\, streaks and vortex clusters) in different wall-normal locations\, suggesting that this new framework provides a more comprehensive way to study turbulence. We also discuss the application of deep reinforcement learning (DRL) to discover active-flow-control strategies for turbulent flows\, including turbulent channels\, three-dimensional cylinders and turbulent separation bubbles. In all the cases\, the discovered DRL-based strategies significantly outperform classical flow-control approaches. We conclude that DRL has tremendous potential for drag reduction in a wide range of complex turbulent-flow configurations. \n 
URL:https://micde.umich.edu/event/fsml-lecture-series-ricardo-vinuesa-identifying-coherent-structures-and-controlling-turbulent-flows-through-deep-learning/
LOCATION:2210 Lurie Engineering Center\, 1221 Beal Ave\, Ann Arbor\, MI\, 48105
CATEGORIES:Engineering,FSML,Science
ATTACH;FMTTYPE=image/png:https://micde.umich.edu/wp-content/uploads/2025/02/Ricardo-Vinuesa.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20250131T120000
DTEND;TZID=America/Detroit:20250131T130000
DTSTAMP:20250127T193042Z
CREATED:20250127T173918Z
LAST-MODIFIED:20250127T193042Z
UID:10000799-1738324800-1738328400@micde.umich.edu
SUMMARY:FSML Lecture Series - Panos Stinis: When big neural networks are not enough: physics\, multi-fidelity and kernels
DESCRIPTION:Zoom link \nBio: Panos Stinis specializes in scientific computing with application interests in model reduction of complex systems\, multiscale modeling\, uncertainty quantification\, and machine learning. He studied aeronautical engineering at the Technical University of Athens\, Greece. He earned his PhD in applied mathematics in 2003\, from Columbia University in New York and began his career as a postdoctoral fellow at Lawrence Berkeley National Laboratory and the Stanford Center for Turbulence Research. In 2008\, he became a faculty member at the Mathematics Department at the University of Minnesota. He moved to the Pacific Northwest National Laboratory in 2014\, where he is currently leading the Computational Mathematics group. \nWhen big neural networks are not enough: physics\, multi-fidelity and kernels\nAbstract: Modern machine learning has shown remarkable promise in multiple applications. However\, brute force use of neural networks\, even when they have huge numbers of trainable parameters\, can fail to provide highly accurate predictions for problems in the physical sciences. We present a collection of ideas about how enforcing physics\, exploiting multi-fidelity knowledge\, and the kernel representation of neural networks can lead to a significant increase in efficiency and/or accuracy. Various examples are used to illustrate the ideas. \n 
URL:https://micde.umich.edu/event/fsml-lecture-series-panos-stinis-when-big-neural-networks-are-not-enough-physics-multi-fidelity-and-kernels/
LOCATION:2004 Lay Auto Lab
CATEGORIES:Engineering,FSML,Science
ATTACH;FMTTYPE=image/png:https://micde.umich.edu/wp-content/uploads/2025/01/Panos-Stinis.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20241206T120000
DTEND;TZID=America/Detroit:20241206T130000
DTSTAMP:20241204T144214Z
CREATED:20241011T181202Z
LAST-MODIFIED:20241204T144214Z
UID:10000779-1733486400-1733490000@micde.umich.edu
SUMMARY:FSML Lecture Series - Anoushka Bhutani: Foundation Model for Molecular Design
DESCRIPTION:Zoom link \nBio: Anoushka is a third-year PhD student in Prof. Venkat Viswanathan’s group at the University of Michigan. Her research interests include machine learning for materials design and electrochemical battery modeling. \nFoundation Model for Molecular Design\nAbstract: The paradigm of molecular machine learning for material screening has accelerated material development cycles\, improved efficiency\, and reduced costs. However\, current state-of-the-art molecular property prediction models still require labeled training data generated using wet-lab experiments or Density Functional Theory (DFT) calculations. Their utility is limited by the scarcity and heterogeneity of labeled materials datasets. Foundation models (FMs) offer a solution to this: these models use self-supervised pre-training strategies to leverage unlabeled datasets and learn representations of data that can be applied to downstream tasks. Large unlabeled datasets of billions of synthesizable molecules are readily available. Prior attempts to train FMs for molecular property prediction demonstrate promise; however\, equivariant geometric models trained using supervised learning are still more accurate. This can be attributed to the fact that foundation models are extremely expensive to train and can be difficult to interpret; they require huge computing budgets\, complex distributed computing techniques\, and extensive hyperparameter searches. Our work addresses these challenges on three fronts: (1) we have prototyped a scalable workflow for distributed training of molecular foundation models (2) we have trained large foundation models using this workflow which demonstrates state-of-the-art molecular property prediction capabilities across several benchmarks\, and (3) we have applied model interpretability strategies such as the attention visualization to shed insight on molecular structure relationships learn by the transformer. \n 
URL:https://micde.umich.edu/event/workshop-seminaranoushka-bhutani-foundation-model-for-molecular-design/
LOCATION:2636 GGBA\, 2350 Hayward St\, Ann Arbor\, MI\, United States
CATEGORIES:Computational Science,Engineering,FSML,Graduate School,Graduate Students,Michigan Engineering,Rackham,Research,Science
ATTACH;FMTTYPE=image/png:https://micde.umich.edu/wp-content/uploads/2024/10/Copy-of-MICDE-2022-2023-Fellowship-Portraits.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20241115T120000
DTEND;TZID=America/Detroit:20241115T130000
DTSTAMP:20260522T154425Z
CREATED:20241012T182154Z
LAST-MODIFIED:20260522T154425Z
UID:10000785-1731672000-1731675600@micde.umich.edu
SUMMARY:FSML Lecture Series - Hongfan Chen: Global Geomagnetic Perturbation Forecasting with Quantified Uncertainty using Deep Gaussian Process
DESCRIPTION:Zoom link \nBio: Hongfan Chen is a third-year PhD student in the Department of Mechanical Engineering at the University of Michigan. His research interests include data assimilation\, uncertainty quantification\, and machine learning applications in space weather. \nGlobal Geomagnetic Perturbation Forecasting with Quantified Uncertainty Using Deep Gaussian Process\nAbstract: Accurately predicting the horizontal component of the ground magnetic field perturbation (dBH)\, as a proxy for Geomagnetically Induced Currents (GICs)\, is crucial for estimating the impact of geomagnetic storms and remains a topic under active investigation. The current state-of-the-practice Geospace model is computationally expensive for fine-grid global simulations while existing machine learning methods consistently tend to underestimate dBH. Additionally\, these models either lack uncertainty quantification (UQ) or provide UQ that lacks calibration. In this work\, as part of the NextGen SWMF project funded by NSF\, we develop a data-driven\, grid-free global model using deep Gaussian process (DGP)\, a Bayesian non-parametric approach that forecasts the dBH for the full surface of Earth with calibrated uncertainty. The model uses solar wind measurements and the Dst index as input\, and it is trained based on ground magnetometer station data provided by SuperMAG over the period 1995-2022. The model’s predictions are evaluated based on the Heidke skill score (HSS) for a total of 22 geomagnetic storms in 2015. We further test the model on the 2024 May 10-12 storm. The results demonstrate that our model outperforms the state-of-the-art model\, with predictions exhibiting high accuracy in mid-latitudes and high-latitude regions in the northern hemisphere. \n 
URL:https://micde.umich.edu/event/lecture-discussionsciml-lecture-series-8/
LOCATION:2636 GGBA\, 2350 Hayward St\, Ann Arbor\, MI\, United States
CATEGORIES:Engineering,FSML,Science
ATTACH;FMTTYPE=image/jpeg:https://micde.umich.edu/wp-content/uploads/2024/10/Hongfan_Chen.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20241101T120000
DTEND;TZID=America/Detroit:20241101T130000
DTSTAMP:20241106T181820Z
CREATED:20241012T182153Z
LAST-MODIFIED:20241106T181820Z
UID:10000784-1730462400-1730466000@micde.umich.edu
SUMMARY:FSML Lecture Series - Nicholas Galioto: Discovery of Cellular Reprogramming Methodology Through Single-cell Foundation Models
DESCRIPTION:Zoom link \nBio: Nick Galioto is a second-year postdoctoral research fellow in the Department of Computational Medicine and Bioinformatics at the University of Michigan (UM). He received his PhD at UM in aerospace engineering in 2023 under the advising of Alex Gorodetsky and remained in the lab for an additional year as a postdoc. In the Gorodetsky lab\, Nick researched how to use stochastic models of dynamical systems to improve system identification. Now\, Nick works in the Rajapakse lab researching how to create data-driven models of the dynamics of cell reprogramming. \n  \nDiscovery of Cellular Reprogramming Methodology Through Single-cell Foundation Models\n  \nAbstract: Cell reprogramming\, the transformation of a cell from one cell type to another through the introduction of exogenous transcription factors (TFs)\, is a rapidly developing research area that could lead to groundbreaking therapeutic technologies in areas such as tissue regeneration\, disease modeling\, and personalized medicine. However\, many challenges still exist that obstruct its practical viability. Discovering which TFs induce reprogramming requires a combinatorial search\, and testing a single candidate set of TFs experimentally can cost tens of thousands of dollars and take multiple months. Moreover\, even when an effective set of TFs is known\, cell conversion efficiency lies only around 5%. Faced with these challenges\, researchers have developed computational surrogate models to rapidly explore the TF search space at a fraction of the cost of wet lab experimentation. Unfortunately\, these models have seen limited success in practice due to the difficulty of capturing the complex gene-gene interactions within the cell\, most of which are still not well understood. With the recent high-profile rise of transformer-based foundation models for natural language\, researchers are now turning to the transformer to push past\, current performance limitations in a wide range of digital biology tasks\, including cell reprogramming. Of particular interest in these models is the attention mechanism\, which is potentially well-suited for capturing long-range gene-gene interactions at a higher fidelity than previously possible. In this talk\, I will describe how the transformer architecture has been adapted for cellular biology and analyze the utility of one such model\, Geneformer\, in identifying TFs for cell reprogramming. Specifically\, I will present the results of an in silico perturbation experiment for reprogramming fibroblast cells to hematopoietic stem cells and compare the outcomes to experimental results found in the literature. I will conclude the talk with a discussion of the drawbacks and limitations of the Geneformer model and provide an assessment of what will be needed in the future for digital biology to fully reap the benefits of large-scale foundation models.
URL:https://micde.umich.edu/event/lecture-discussionsciml-lecture-series-7/
LOCATION:Walter E Lay Auto Lab – 2052
CATEGORIES:Engineering,FSML,Science
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20241018T120000
DTEND;TZID=America/Detroit:20241018T130000
DTSTAMP:20241104T135643Z
CREATED:20241012T182153Z
LAST-MODIFIED:20241104T135643Z
UID:10000783-1729252800-1729256400@micde.umich.edu
SUMMARY:FSML Lecture Series: Domain decomposition and coupling data-driven models of fluid flows by Christopher Wentland\, Sandia National Labs
DESCRIPTION:Zoom link \n \nAbstract: Simulating complex physical systems often requires joining non-uniform subsystems\, which may be characterized by different geometries or mesh topologies. Coupling these separate subsystems often relies on time-intensive meshing workflows or empirical coupling models\, which may not generalize well across all operational regimes. The Schwarz alternating method proposes to overcome these issues\, establishing an effective domain decomposition framework that allows for the coupling of arbitrary geometries. This talk presents a brief history of Schwarz-based coupling work at Sandia National Laboratories\, along with recent work on combining the Schwarz alternating method with data-driven modeling approaches\, namely projection-based reduced order models (PROMs) and operator inference. This approach can generate surrogates that are capable of simulating advection-dominated fluid flows with higher accuracy and lower cost than comparable monolithic models\, aiding analysis in many-query applications such as uncertainty quantification and engineering design. Several nuances of the Schwarz algorithm and their impacts on model performance are explored\, specifically non-overlapping decompositions and PROM hyper-reduction under domain decomposition. A look into ongoing Schwarz coupling work at Sandia discusses existing challenges and efforts to apply this approach to relevant engineering problems.
URL:https://micde.umich.edu/event/lecture-discussionsciml-lecture-series-6/
LOCATION:2636 GGBA\, 2350 Hayward St\, Ann Arbor\, MI\, United States
CATEGORIES:Engineering,FSML,Science
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Detroit:20241004T120000
DTEND;TZID=America/Detroit:20241004T130000
DTSTAMP:20241104T140255Z
CREATED:20241002T220654Z
LAST-MODIFIED:20241104T140255Z
UID:10000774-1728043200-1728046800@micde.umich.edu
SUMMARY:FSML Lecture Series: Tokenization for Chemistry by Alex Wadell\, University of Michigan
DESCRIPTION:Alex Wadell is a PhD candidate at the University of Michigan Department of Mechanical Engineering. \nTokenization for Chemistry\nMolecular Foundation Models are emerging as a powerful tool for molecular design\, material science\, and cheminformatics. By leveraging the transformer architecture\, these models attempt to learn the language of chemistry and discover robust molecular embeddings. However\, current models are constrained by tokenizers that fail to capture the full breadth of chemical space or even the periodic table of elements. In his talk\, Alex will introduce smirk\, a new tokenizer for molecular foundation models that can represent the entirety of the OpenSMILES specification. We’ll also discuss performance metrics for tokenizers and the results of Alex’s systematic evaluation of thirteen chemistry-specific tokenizers using N-gram language models as a low-cost proxy for transformer models. \nIf you are unable to attend in person but are interested\, please feel free to join virtually. \nJoin Zoom Meeting\nhttps://umich.zoom.us/j/97823527756?pwd=H01BbvtuG5q02Wzb8LJvhUnvijlAIe.1\nMeeting ID: 978 2352 7756\nPasscode: 2024
URL:https://micde.umich.edu/event/lecture-discussionsciml-lecture-series/
LOCATION:Walter E Lay Auto Lab – 2052
CATEGORIES:FSML,Micde,Sciml
ATTACH;FMTTYPE=image/png:https://micde.umich.edu/wp-content/uploads/2024/10/SciML-Lectures-e1727980667262.png
END:VEVENT
END:VCALENDAR