Oden Institute Past Seminars
Inperson seminars are held Tuesdays and Thursdays in POB 6.304 from 3:305:00 pm, unless otherwise noted. Zoom seminars usually occur during the same time period. Speakers include scientists, researchers, visiting scholars, potential faculty, and Oden Institute/UT Faculty or staff. Everyone is welcome to attend. Refreshments are served at inperson seminars at 3:15 pm.
Thursday, May 27
Quantum mechanical embeddings of classical dynamical systems
Thursday, May 27, 2021 from 3:30PM to 5PM  Zoom Meeting

Additional Information
Hosted by PerGunnar J. Martinsson and Rachel Ward
Sponsor: Oden Institute Virtual Seminar
Speaker: Dimitris Giannakis
Speaker Affiliation: Associate Professor, Mathematics, Center for Atmosphere Ocean Science (CAOS), Courant Institute of Mathematical Sciences, NYU

Abstract
We present a framework for simulating a classical dynamical system by a finitedimensional quantum system amenable to implementation on a quantum computer. The framework is based on a quantum feature map for representing classical states by density operators (quantum states) on a reproducing kernel Hilbert space (RKHS), H. Simultaneously, a mapping is employed from classical observables into selfadjoint operators on H such that quantum mechanical expectation values are consistent with pointwise function evaluation. Meanwhile, quantum states and observables on H evolve under the action of a unitary group of Koopman operators in a consistent manner with classical dynamical evolution. To achieve quantum parallelism, the state of the quantum system is projected onto a finiterank density operator on a 2^Ndimensional tensor product Hilbert space associated with N qubits. In this talk, we describe this "quantum compiler" framework, and illustrate it with applications to lowdimensional dynamical systems. In addition, we discuss a related quantum approach for data assimilation of partially observed systems along with applications to climate dynamics.
Bio
Dimitris Giannakis is an Associate Professor of Mathematics at the Courant Institute of Mathematical Sciences, New York University. He is also affiliated with Courant's Center for Atmosphere Ocean Science (CAOS). He received BA and MSci degrees in Natural Sciences from the University of Cambridge in 2001, and a PhD degree in Physics from the University of Chicago in 2009. Giannakis' current research focus is at the interface between operatortheoretic techniques for dynamical systems and machine learning. His recent work includes the development of techniques for coherent pattern extraction, statistical forecasting, and data assimilation based on datadriven approximations of Koopman operators of dynamical systems. He has worked on applications of these tools to atmosphere ocean science, fluid dynamics, and molecular dynamics. Giannakis received a Young Investigator Program award from the Office of Naval Research in 2016 and a Vannevar Bush Faculty Fellowship in 2021.
Thursday, May 20
Scientific computing paradigms in scaling data science and network science
Thursday, May 20, 2021 from 3:30PM to 5PM  Zoom Meeting

Additional Information
Hosted by PerGunnar J. Martinsson
Sponsor: Oden Institute Virtual Seminars
Speaker: David Gleich
Speaker Affiliation: Associate Professor, Computer Science Department, Purdue University

Abstract
Common paradigms to understand and scale computational datadriven analysis include
 scaling computational resources
 improving algorithms
 improving models
or a combination of these. I will present an overview of contributions from my research team in these areas in scenarios that span data science, network science, scientific computing, and combinatorial scientific computing. We will also take a deeper look into our recent and ongoing research in higherorder methods for networks and hypergraphs where there are deep interactions between the algorithms and models.Helpful Overview of Higherorder Networks:
https://sinews.siam.org/DetailsPage/higherordernetworkanalysistakesofffueledbyoldideasandnewdataCodes:
https://github.com/kfoynt/LocalGraphClustering
https://github.com/MengLiuPurdue/LHQDBio
David Gleich is the Jyoti and Aditya Mathur Associate Professor in the Computer Science Department at Purdue University whose research is on novel models and fast largescale algorithms for datadriven scientific computing including scientific data analysis, bioinformatics, and network analysis. He is committed to making software available based on this research and has written software packages such as MatlabBGL with thousands of users worldwide. Gleich has received a number of awards for his research including a SIAM Outstanding Publication prize (2018), a Sloan Research Fellowship (2016), an NSF CAREER Award (2011), the John von Neumann postdoctoral fellowship at Sandia National Laboratories in Livermore CA (2009). His research is funded by the NSF, DOE, DARPA, and NASA. For more information, see his website: https://www.cs.purdue.edu/homes/dgleich/
Friday, May 7
Adaptive methods for the simulation of diffusion and fluid flow in complex geometries
Friday, May 7, 2021 from 10AM to 11AM  Zoom Meeting

Additional Information
Hosted by Anna Yesypenko
Sponsor: Oden Institute Virtual Seminar  Babuška Forum series
Speaker: Leslie Greengard
Speaker Affiliation: Director of the Center for Computational Mathematics at the Flatiron Institute, a division of the Simons Foundation

Abstract
We will review the state of the art in integral equation methods for the solution of the heat equation and fluid flow in moving geometries. With suitable fast algorithms, such methods achieve optimal complexity and, in the homogeneous case, require the discretization of the spacetime boundary alone. They achieve high order accuracy with suitable quadratures and are straightforward to implement adaptively in spacetime. We will discuss applications to biophysical modeling, reactiondiffusion systems, and incompressible fluid dynamics.
Bio
Leslie Greengard received his B.A. degree in Mathematics from Wesleyan University in 1979, and his M.D. and Ph.D. degrees from Yale University in 1987. From 19871989 he was an NSF Postdoctoral Fellow at Yale University and at the Courant Institute of Mathematical Sciences, NYU, where he is a member of the faculty. He served as the Director of the Courant Institute from 20062011. He is presently Director of the Center for Computational Mathematics at the Flatiron Institute, a division of the Simons Foundation. Greengard, together with V. Rokhlin, developed the Fast Multipole Method (FMM) for problems in gravitation, electrostatics and electromagnetics. Much of Greengard’s research has been aimed at the development of highorder accurate integral equation methods for partial differential equations in complex geometry. He is a member of the National Academy of Sciences, the National Academy of Engineering and the American Academy of Arts and Sciences.(The Babuška Forum series was started by Professor Ivo Babuška several years ago to expose students to interesting and curious topics relevant to computational engineering and science with technical content at the graduate student level (i.e. the focus of the lectures is on main ideas with some technical content). Seminar credit is given to those students who attend.)
Friday, May 7

Additional Information
Hosted by Shane McQuarrie
Sponsor: Oden Institute Virtual Seminar  CSEM Student Forum series  Blitz Presentations
Speaker: 1: Allison Torsey, 2: Samuel Majors, 3: Jacob Badger
Speaker Affiliation: 2nd year Ph.D. candidates, CSEM, Oden Institute, UT Austin

Abstract
Three CSEM Graduate students summarize some of their current work.
Bio
1: Allison Torsey is a secondyear PhD student in Dr. David Paydarfar lab group in the Dell Medical School. Her work mainly revolves around understanding apneas in premature infants and modeling lung dynamics.2: Samuel Majors is a secondyear PhD student with a background in aerospace engineering. He is advised by Dr. Karen Willcox, and works on model reduction for supersonic and hypersonic flows in collaboration with Sandia National Laboratories.
3: Jacob Badger is a secondyear PhD student working with Dr. Leszek Demkowicz. He received his BS in Mathematics and Mechanical Engineering from Brigham Young University, where he was introduced to computational modeling through work on curvedfold origami.
(The CSEM Student Forum is a seminar series given by current CSEM graduate students to their peers. The aim of the forum is to expose students to each other's research, encourage collaboration, and provide opportunities to practice presentation skills. First and secondyear CSEM students receive seminar credit for attending. Blitz presentations are given by multiple secondyear students who will summarize some of their current work and the work within their research groups.)
Thursday, May 6
Quantum numerical linear algebra
Thursday, May 6, 2021 from 3:30PM to 5PM  Zoom Meeting

Additional Information
Hosted by PerGunnar J. Martinsson
Sponsor: Oden Institute Virtual Seminar
Speaker: Lin Lin
Speaker Affiliation: Associate Professor, Department of Mathematics, University of California, Berkeley

Abstract
The two "quantum supremacy" experiments (by Google in 2019 and by USTC in 2020, respectively) have brought quantum computation to the public's attention. In this talk, I will discuss how to use a quantum computer to solve linear algebra problems. I will start with a toy linear system Ax=b, where A is merely a 2 x 2 matrix. I will then talk about some recent progress of quantum linear system solvers, eigenvalue problems, and a proposal for the quantum LINPACK benchmark.
Bio
Lin Lin received his B.S. degree in Computational Mathematics from Peking University in 2007, and Ph.D. degree in Applied and Computational Mathematics from Princeton University in 2011, advised by Professor Weinan E and Professor Roberto Car. His research focuses on the development of efficient and accurate numerical methods for electronic structure calculations, with broad applications in quantum chemistry, quantum physics and materials science. He is now an associate professor in the Department of Mathematics at UC Berkeley, a faculty scientist at Berkeley Lab’s Mathematics Group within the Computational Research Division, and a mathematician within Berkeley Lab's Center for Advanced Mathematics for Energy Research Applications (CAMERA). He received the Sloan Research Fellowship (2015), the National Science Foundation CAREER award (2017), the Department of Energy Early Career award (2017), the inaugural SIAM Computational Science and Engineering (CSE) early career award (2017), the Presidential Early Career Awards for Scientists and Engineers (PECASE) (2019), and the ACM Gordon Bell Prize (Team, 2020).
Friday, Apr 30
Software development in the sixth epoch of distributed computing
Friday, Apr 30, 2021 from 10AM to 11AM  Zoom Meeting

Additional Information
Hosted by Anna Yesypenko
Sponsor: Oden Institute Virtual Seminar  Babuška Forum series
Speaker: Tim Mattson
Speaker Affiliation: Senior principal engineer, Intel

Abstract
Amin Vahdat, in a talk that has gone viral, described the five epochs of distributed computing (https://www.youtube.com/watch?v=Am_itCzkaE0). It’s a great talk, but I disagree with him on one key point. He thinks we are early in the fifth Epoch. I say we entered the fifth Epoch several years ago and we are on the verge of the next Epoch … the sixth Epoch of distributed computing.
In this talk I will very briefly outline the five Epochs of distributed computing and then shift to the future and the sixth Epoch. This Epoch emerges when we bring next generation networking technology into our distributed computing systems so the time for one hop on the network is on par with the time for a memory reference in DRAM (Distributed Random Access Memory).
This innovation is coming in the nottoodistant future. It will fundamentally change how highperformance computing applications project into the cloud. We need to start thinking NOW about how we will develop software in the sixth Epoch. I will suggest one approach for programming in the sixth Epoch, but the ideas are speculative and therefore alternatives abound. To that end, I hope this talk launches an aggressive, and hopefully interesting, dialog about software development in the sixth epoch of distributed computing.
Bio
Tim Mattson is a parallel programmer obsessed with every variety of science (Ph.D. Chemistry, UCSC, 1985). He is a senior principal engineer in Intel’s parallel computing lab. Tim has been with Intel since 1993 and has worked with brilliant people on great projects including: (1) the first TFLOP computer (ASCI Red), (2) MPI, OpenMP and OpenCL, (3) two different research processors (Intel's TFLOP chip and the 48 core SCC), (4) Data management systems (Polystore systems and Arraybased storage engines), and (5) the GraphBLAS API for expressing graph algorithms as sparse linear algebra. Tim has well over 150 publications including five books on different aspects of parallel computing, the latest (Published November 2019) titled “The OpenMP Common Core: making OpenMP Simple Again”.(The Babuška Forum series was started by Professor Ivo Babuška several years ago to expose students to interesting and curious topics relevant to computational engineering and science with technical content at the graduate student level (i.e. the focus of the lectures is on main ideas with some technical content). Seminar credit is given to those students who attend.)
Thursday, Apr 29
Humanmachine interaction models in roboadvising
Thursday, Apr 29, 2021 from 3:30PM to 5PM  Zoom Meeting

Additional Information
Hosted by Leszek Demkowicz
Sponsor: Oden Institute Virtual Seminar
Speaker: Thaleia Zariphopoulou
Speaker Affiliation: Chair in Mathematics, V. F. Neuhaus Centennial Professorship in Finance, and a member of the Affiliated Faculty at the Oden Institute, UT Austin

Abstract
In my talk, I will introduce a family of humanmachine interaction (HMI) models in optimal portfolio construction (roboadvising) and, more generally, in optimal resource allocation and risk management. Modeling difficulties stem from the limited ability to quantify the human’s risk preferences and describe their evolution, but also from the fact that the stochastic environment, in which the machine optimizes, adapts to realtime incoming information that is exogenous to the human. Furthermore, the human’s risk preferences and the machine’s states may evolve at different scales. This interaction creates an adaptive cooperative game with both asymmetric and incomplete information exchange between the two parties. As a result, challenging questions arise on, among others, how frequently the two parties should communicate, what information can the machine accurately detect, infer and predict, how the human reacts to exogenous events, how to improve the interlinked reliability between the human and the machine, how to measure the performance of the interactive system, and others. Such HMI models give rise to new, nonstandard optimization problems that combine a new class (forward/ill posed) adaptive stochastic control, non zerosum stochastic differential games, timeinconsistency, optimal stopping, multiscales and learning.
Bio
Thaleia Zariphopoulou is the holder of the Presidential Chair of Mathematics and the V.F. Neuhaus Professorship of Finance at the University of Texas at Austin. Previously, she was the Laun Professor at the University of Wisconsin, Madison and from 20092012, the first holder of the OxfordMan Chair in Quantitative Finance at the University of Oxford. She currently is a Professor in the Department of Mathematics in the College of Natural Sciences and a Professor in the Department of Information, Risk, and Operations Management at the Red McCombs School of Business.Her area of expertise is Financial Mathematics and Stochastic Optimization. She has published extensively in the areas of investments and valuation in incomplete markets, and introduced novel approaches to indifference valuation and dynamic risk preferences.
She has served very actively the community of Financial Mathematics. She sits on the editorial board of eleven academic journals and monograph series, and she is the Editor of the SIAM Series in Financial Mathematics. She has served in various prize committees and panels. She has also been the ViceChair (20072010) of the SIAG Activity Group in Financial Mathematics and Engineering, and has served as VicePresident (20042006) and President (20062008) of the Bachelier Finance Society.
In 2012, she was elected SIAM Fellow and in 2014, she was an invited speaker at the International Congress of Mathematicians in Seoul.
Friday, Apr 23
Stochastic Gradient Descent with Adaptive Stepsizes: from practice to theory
Friday, Apr 23, 2021 from 10AM to 11AM  Zoom Meeting

Additional Information
Hosted by Anna Yesypenko
Sponsor: Oden Institute Virtual Seminar  Babuška Forum series
Speaker: Rachel Ward
Speaker Affiliation: Professor, Mathematics, UT Austin and Oden Institute

Abstract
Stochastic Gradient Descent (SGD) is an increasingly popular optimization algorithm for a variety of largescale learning problems, due to its computational efficiency and ease of implementation. In particular, SGD is the standard algorithm for training neural networks. Still, there remains is a wide gap between the setting where SGD theoretical guarantees and the setting where SGD is most effective and useful in practice. Aiming to reduce this gap, we present first theoretical guarantees for an "adaptive learning rate" SGD algorithm (AdaGrad). Such algorithms are valuable in practice for making SGD behavior less sensitive to the choice of stepsizes, but have proven difficult to analyze theoretically until now due to their nonlinear dynamics.
Bio
Rachel Ward is the W.A. "Tex” Moncrief Distinguished Professor in Computational Engineering and Sciences — Data Science and Professor of Mathematics at UT Austin. She is recognized for her contributions to sparse approximation, stochastic optimization, and numerical linear algebra. Prior to joining UT Austin in 2011, Dr. Ward received the PhD in Computational and Applied Mathematics at Princeton in 2009 and was a Courant Instructor at the Courant Institute, NYU, from 20092011. Among her awards are the Sloan research fellowship, NSF CAREER award, and the 2016 IMA prize in mathematics and its applications.(The Babuška Forum series was started by Professor Ivo Babuška several years ago to expose students to interesting and curious topics relevant to computational engineering and science with technical content at the graduate student level (i.e. the focus of the lectures is on main ideas with some technical content). Seminar credit is given to those students who attend.)
Friday, Apr 23
Asynchronous Marching for Eikonal Equations
Friday, Apr 23, 2021 from 2PM to 3PM  Zoom Meeting

Additional Information
Hosted by Shane McQuarrie
Sponsor: Oden Institute Virtual Seminar  CSEM Student Forum series
Speaker: Ian Henriksen
Speaker Affiliation: Ph.D. Candidate, CSEM, Oden Institute, UT Austin

Abstract
Numerical solutions to the Eikonal equation are computed using variants of the fast marching method, the fast sweeping method, and the fast iterative method. These algorithms differ primarily in the ordering constraints imposed on the intermediate states of the algorithms. In this talk we will discuss why existing Eikonal solvers may produce different results despite using the same update scheme and demonstrate techniques to address these discrepancies. Once these numerical concerns are addressed, it becomes possible to apply modern concurrent priority scheduling techniques to Eikonal solvers and run the problem fully asynchronously. Doing so results in good parallel performance for a problem from seismology.
Bio
Ian Henriksen is a PhD candidate in the Oden Institute working with Keshav Pingali. Prior to coming to UT Austin he worked at Anaconda Inc. as an open source developer. He has bachelor's and master's degrees in mathematics from Brigham Young University and has served as a developer and maintainer for the SciPy, DyND, Galois, and Parla open source projects.
Friday, Apr 16
Finding structure with randomness
Friday, Apr 16, 2021 from 11AM to 12PM  Zoom Meeting

Additional Information
Hosted by Anna Yesypenko
Sponsor: Oden Institute Virtual Seminar  Babuška Forum series
Speaker: Joel Tropp
Speaker Affiliation: Professor, CalTech

Abstract
Over the last 20 years, randomized algorithms have revolutionized the field of matrix computations. These new methods can efficiently and robustly solve huge linear algebra problems that were previously inaccessible.
This talk introduces the randomized singular value decomposition (SVD) algorithm, perhaps the most widely used method that has emerged from this research program. The randomized SVD algorithm supports largescale linear regression, principal component analysis, proper orthogonal decomposition, and many other methods for data reduction and summarization. The talk offers a highlevel view of how randomness facilitates the SVD computation, the kinds of theoretical guarantees it allows, and some applications in science and engineering.
For more information, see the papers arXiv:0909.4061 and arXiv:2002.01387.
Bio
Joel A. Tropp is Steele Family Professor of Applied and Computational Mathematics at Caltech. His research centers on data science, applied mathematics, numerical algorithms, and random matrix theory. He attained the Ph.D. degree in Computational Applied Mathematics at the University of Texas at Austin in 2004, and he joined Caltech in 2007. Prof. Tropp won the PECASE in 2008, and he was recognized as a Highly Cited Researcher in Computer Science each year from 2014–2018. He is cofounder and Section Editor of the SIAM Journal on Mathematics of Data Science (SIMODS), and he was cochair of the inaugural 2020 SIAM Conference on the Mathematics of Data Science. Prof. Tropp was elected SIAM Fellow in 2019 and IEEE Fellow in 2020.(The Babuška Forum series was started by Professor Ivo Babuška several years ago to expose students to interesting and curious topics relevant to computational engineering and science with technical content at the graduate student level (i.e. the focus of the lectures is on main ideas with some technical content). Seminar credit is given to those students who attend.)