MAZIARRAISSI · EDUCATION Ph.D.inAppliedMathematics&Statistics,andScientificComputation M.A.inEconomics UniversityofMarylandCollegePark 2013–2016 ‰CollegePark,MD

MAZIARRAISSIAssistant Professor of AppliedMathematics, University of Colorado Boulder½ Engineering Center, ECOT 332, 526 UCB, Boulder, CO 80309-0526 R [email protected] [email protected] Ó +1 (303) 735-4434 Ó +1 (202) 812-5606® https://www.colorado.edu/amath/maziar-raissi ® https://maziarraissi.github.io/� https://github.com/maziarraissi/ ¯ https://www.linkedin.com/in/mraissi/7 https://twitter.com/MaziarRaissi � https://scholar.google.com/citations?user=dCdmUaYAAAAJ&hl=en

EXPERIENCEAssistant ProfessorUniversity of Colorado BoulderDepartment of AppliedMathematics� January 2020 –Ongoing ½ Boulder, CO• Doing research at the intersection of ProbabilisticMachine Learning, Deep Learning, and Data-driven Scientific Computing(40%).

• Teaching Introduction to Data Science using GitHub, R, and Python (40%).• Serving on the Undergradate Committee and the Statistics &Data Science Committee (20%).• The AppliedMathematics program at the University of Colorado Boulder is ranked 14 among the best graduate schools inthe US.

Senior Software EngineerNVIDIA Corporation� December 2018 – January 2020 ½ Santa Clara, CA• Initiated a data-efficient deep learning product demoed by Jensen Huang (NVIDIA’s Founder and CEO) at SC19, theInternational Conference for High Performance Computing, Networking, Storage, and Analysis.

• Learned about Ray Tracing, DLSS: Deep Learning Super Sampling, Digital Humans, Omniverse, 5G,MassiveMIMO,mmWaves, Cloud Computing, Kubernetes, Helm, CI/CD, Jenkins, VLSI, Volta, Turing, DGX, EGX: Platform for EdgeComputing, NGC: Platform for Cloud Computing, GFN: GeForce NOW, PhysX, Clara SDK, Genomics, IoT: Internet ofThings, Digital Twins, IVA: Intelligent Video Analytics, Autonomous Vehicles, Xavier, Pegasus, Orin, Drive AV, Drive IX,Drive OS, Drive Sim, Constellation, Jarvis: Multimodal AI SDK, NeMo: NeuralModules, Robotics, Isaac SDK, etc.

Research Assistant ProfessorBrownUniversityDivision of AppliedMathematics� August 2017 –December 2018 ½ Providence, RI• Led two grants (DARPA and AFOSR) onmachine learning, deep learning, and data-driven scientific computing.• Releasedmore that 15 open-source C++, Cuda, Python andMatlab projects on GitHub.

Postdoctoral Research AssociateBrownUniversityDivision of AppliedMathematics� January 2016 – August 2017 ½ Providence, RI• Conceived the original ideas of physics-informedmachine learning and deep learning.• The appliedmathematics program at BrownUniversity is ranked 4 among the best graduate schools in the US.

Quantitative Research AssociateWorld Bank GroupInternational Finance Corporation (IFC)TreasuryQuantitative Analysis� February 2015 – August 2015 ½ Washington, DC• Developed C++ codes tomerge Summit, a multi-currency professional investment accounting system, and an in-housederivative pricing toolbox.

• Workedwith Bloomberg Terminal, Numerix, and Summit on a daily basis.

EDUCATIONPh.D. in AppliedMathematics & Statistics, and Scientific ComputationM.A. in EconomicsUniversity ofMaryland College Park� 2013 – 2016 ½ College Park, MD• Dissertation Topic: Conic Economics• Advisor: DilipMadan• The AppliedMathematics program at the University ofMaryland College Park is ranked 13 among the best graduateschools in the US.

• The Economics program at University ofMaryland College Park is ranked 21 among the best graduate schools in the US.• Grade Point Average: 3.46 out of 4• Courses: AdvancedMacroeconomics I, ComputationalMethods inMacroeconomics, Econometrics I & II & III,Macroeconomic Analysis I & II, Microeconomic Analysis I & II, Numerical Methods in Partial Differential Equations,Seminar in Financial Theory, AdvancedNumerical Analysis

Ph.D. in AppliedMathematicsGeorgeMasonUniversity� 2011 – 2013 ½ Fairfax, VA• Dissertation Topic: Multi-fidelity Stochastic CollocationMethods• Advisor: Padmanabhan Seshaiyer• Grade Point Average: 3.97 out of 4• Courses: AdvancedMethods in AppliedMathematics, Banach Spaces of Analytic Functions, Complex Functions,Computational Analysis of Social Complexities - Agent-based Computing in Social Sciences, Fourier Analysis, Mathematical(Financial) Derivatives, Mathematics of Finite ElementMethod, Numerical Linear Algebra, Stochastic Finite Elements,Stochastic Processes

M.S. in AppliedMathematicsIsfahan University of Technology� 2008 – 2011 ½ Isfahan, Iran• Dissertation Topic: Numerical Continuation of Connecting Orbits ofMaps inMatlab• Advisors: RezaMokhtari and Reza Khoshsiar Ghaziani• Grade Point Average: 17.53 out of 20• Courses: Dynamical Systems I & II, Nonlinear Functional Analysis, Nonlinear PDEs, Numerical Analysis I & II, NumericalSolutions to Ordinary Differential Equations, Numerical Solutions to Partial Differential Equations, Real Analysis

B.S. in AppliedMathematicsUniversity of Isfahan� 2004 – 2008 ½ Isfahan, Iran• Grade Point Average: 17.30 out of 20• Courses: Algebra I & II, Calculus I & II, Data Structures and Algorithms, Database Design, DiscreteMathematics,Fundamentals of Economics, Fundamentals ofMathematics, Linear Algebra I & II, Math Softwares, Mathematical Analysis I& II, Operations Research, Optimization, Partial Differential Equations, Probability and Statistics I & II, AdvancedProgramming, Differential Equations

PUBLICATIONS� Dissertations[1] Maziar Raissi. “Conic Economics”. PhD thesis. University ofMaryland, College Park, 2016. URL:

http://bit.ly/2hkIHZ1.[2] Maziar Raissi. “Multi-fidelity Stochastic Collocation”. PhD thesis. GeorgeMason University, 2013. URL:

http://bit.ly/2xggpcn.

q Journal Articles[3] Maziar Raissi, Alireza Yazdani, and George EmKarniadakis. “Hidden FluidMechanics: Learning Velocity and Pressure

Fields from FlowVisualizations”. In: Science (2020). URL: https://bit.ly/2SyVVpa.[4] Mamikon Gulian, Maziar Raissi, Paris Perdikaris, and George Karniadakis. “Machine Learning of Space-fractional

Differential Equations”. In: SIAM Journal on Scientific Computing 41.4 (2019), A2485–A2509. URL:https://bit.ly/396B3LX.

[5] Maziar Raissi, HessamBabaee, and PeymanGivi. “Deep Learning of Turbulent ScalarMixing”. In: Phys. Rev. Fluids 4 (12Dec. 2019), p. 124501. URL: https://bit.ly/35RYDu3.

[6] Maziar Raissi, HessamBabaee, and George EmKarniadakis. “Parametric Gaussian Process Regression for Big Data”. In:Computational Mechanics 64.2 (2019), pp. 409–416. URL: https://bit.ly/34UsvVg.

[7] Maziar Raissi, Paris Perdikaris, and George E Karniadakis. “Physics-InformedNeural Networks: A Deep LearningFramework for Solving Forward and Inverse Problems Involving Nonlinear Partial Differential Equations”. In: Journal ofComputational Physics 378 (2019), pp. 686–707. URL: https://bit.ly/2SmVsqq.

[8] Maziar Raissi, Niloofar Ramezani, and Padmanabhan Seshaiyer. “On Parameter Estimation Approaches for PredictingDisease Transmission throughOptimization, Deep Learning and Statistical InferenceMethods”. In: Letters inBiomathematics (2019), pp. 1–26. URL: https://bit.ly/2Ss3sXf.

[9] Maziar Raissi. “DeepHidden PhysicsModels: Deep Learning of Nonlinear Partial Differential Equations”. In: Journal ofMachine Learning Research 19.25 (2018), pp. 1–24. URL: http://jmlr.org/papers/v19/18-046.html.

[10] Maziar Raissi and George EmKarniadakis. “Hidden PhysicsModels: Machine Learning of Nonlinear Partial DifferentialEquations”. In: Journal of Computational Physics 357 (2018), pp. 125–141. URL: https://bit.ly/2D7oi4p.

[11] Maziar Raissi, Paris Perdikaris, and George EmKarniadakis. “Numerical Gaussian Processes for Time-Dependent andNonlinear Partial Differential Equations”. In: SIAM Journal on Scientific Computing 40.1 (2018), A172–A198. URL:https://bit.ly/2K5rIXL.

[12] Maziar Raissi and Padmanabhan Seshaiyer. “Application of Local Improvements to Reduced-orderModels to SamplingMethods for Nonlinear PDEswith Noise”. In: International Journal of Computer Mathematics 95.5 (2018), pp. 870–880.URL: http://bit.ly/2jLMsLR.

[13] Paris Perdikaris, Maziar Raissi, Andreas Damianou, Neil D. Lawrence, and George EmKarniadakis. “NonlinearInformation Fusion Algorithms for Data-efficientMulti-fidelityModelling”. In: Proceedings of the Royal Society of LondonA: Mathematical, Physical and Engineering Sciences 473.2198 (2017). URL: http://bit.ly/2w7HJWx.

[14] Maziar Raissi, Paris Perdikaris, and George EmKarniadakis. “Inferring Solutions of Differential Equations using NoisyMulti-fidelity Data”. In: Journal of Computational Physics 335 (2017), pp. 736–746. URL: http://bit.ly/2jMANfP.

[15] Maziar Raissi, Paris Perdikaris, and George EmKarniadakis. “Machine Learning of Linear Differential Equations usingGaussian Processes”. In: Journal of Computational Physics 348 (2017), pp. 683–693. URL: http://bit.ly/2fC9ccs.

[16] Paul Cashin, KamiarMohaddes, Maziar Raissi, andMehdi Raissi. “The Differential Effects of Oil Demand and SupplyShocks on the Global Economy”. In: Energy Economics 44 (2014), pp. 113–134. URL: http://bit.ly/2yrqv88.

[17] Maziar Raissi and Padmanabhan Seshaiyer. “AMulti-fidelity Stochastic CollocationMethod for Parabolic PartialDifferential Equations with Random Input Data”. In: International Journal for Uncertainty Quantification 4.3 (2014). URL:http://bit.ly/2yryAJR.

q Preprints[18] Alireza Yazdani, Maziar Raissi, and George EmKarniadakis. “Systems biology informed deep learning for inferring

parameters and hidden dynamics”. In: bioRxiv (2019). URL: https://bit.ly/2sh3Son.[19] Maziar Raissi. “Forward-Backward Stochastic Neural Networks: Deep Learning of High-dimensional Partial

Differential Equations”. In: arXiv preprint arXiv:1804.07010 (2018). URL: https://arxiv.org/abs/1804.07010.[20] Maziar Raissi, Paris Perdikaris, and George EmKarniadakis. “Multistep Neural Networks for Data-driven Discovery of

Nonlinear Dynamical Systems”. In: arXiv preprint arXiv:1801.01236 (2018). URL: https://arxiv.org/abs/1801.01236.[21] Maziar Raissi and George Karniadakis. “DeepMulti-fidelity Gaussian Processes”. In: arXiv preprint arXiv:1604.07484

(2016). URL: https://arxiv.org/abs/1604.07484.

RESEARCH INTERESTSWithin the field of appliedmathematics, my research interests span the areas of probabilistic machine learning, deeplearning, data-driven scientific computing, multi-fidelity modeling, uncertainty quantification, big data analysis, economics,and finance.5 Watchmy talk: https://bit.ly/2MkU2J2

PATENTS&AWARDS­ Physics Informed LearningMachines

U.S. Provisional Patent Application 6248319,March 29, 2017.3 Office of Provost Fellowship

GeorgeMason University3 Ranked first

National university entrance exam for PhD degree in AppliedMathematics in 2011 among thousands of participants.

RECENT TALKS• Department ofMechanical Engineering, Rice University, February 19, 2020, Houston, TX, USA.• Institute for Pure & AppliedMathematics, UCLA, October 29, 2019, Los Angeles, CA, USA. Video:5• TheMonterey Data Conference, August 8, 2019,Monterey, CA, USA.• Computational Science Research Center, June 1–10, 2019, Beijing, China.• Jack Baskin School of Engineering, University of California Santa Cruz, April 22, 2019, Santa Cruz, CA, USA.• Center ofMathematical Sciences and Applications, Harvard University, April 17, 2019, Cambridge, MA, USA.• School of Earth, Energy & Environmental Sciences, Stanford University, April 1, 2019, Stanford, CA, USA.• Swanson School of Engineering, University of Pittsburgh, February 21, 2019, Pittsburgh, PA, USA.• Department ofMathematics, University of South Carolina, January 24, 2019, Columbia, SC, Canada.• ICERMWorkshop on ScientificMachine Learning, January 28–30, 2019, Providence, RI, USA. Video:5• School ofMathematical and Statistical Sciences, Arizona State University, January 17, 2019, Tempe AZ, USA.• Department of AppliedMathematics, University ofWaterloo, January 7, 2019,Waterloo, ON, Canada.• Department ofMechanical Engineering, Massachusetts Institute of Technology, November 30, 2018, Cambridge, MA, USA.• Department ofMathematics and Statistics, University ofMaryland, October 5, 2018, Baltimore County, MD, USA.• NVIDIA, September 27, 2018, Santa Clara, CA, USA.• Schlumberger-Doll Research Center, September 6, 2018, Cambridge, MAUSA.• The 13thWorld Congress in ComputationalMechanics, July 22–27, 2018, New York City, NY, USA.• SIAMAnnualMeeting, July 9–13, 2018, Portland, OR, USA.• School of Computational Science and Engineering, Georgia Tech, February 8, 2018, Atlanta, GA, USA.• DARPA EQUiPS PI ReviewMeeting, February 12–13, 2018, Arlington, VA, USA.• School of Natural Sciences, University of California, January 26, 2018,Merced, CA, USA.• Michigan Institute for Computational Discovery and Engineering, University ofMichigan, December 4, 2017, Ann Arbor,MI, USA.

• Schlumberger-Doll Research Center, October 5, 2017, Cambridge, MAUSA.• Department ofMechanical Engineering,Massachusetts Institute of Technology, September 14, 2017, Cambridge,MA, USA.• DARPA EQUiPS PI ReviewMeeting, August 16–18, 2017, Arlington, VA, USA.• SIAMAnnualMeeting, July 10–14, 2017, Pittsburgh, PA.• ICERMWorkshop on Probabilistic Scientific Computing, June 5–9, 2017, Providence, RI, USA. Video:5• DARPA EQUiPS PI ReviewMeeting, March 28–30, 2017, Austin, TX, USA.• DARPA EQUiPS PI ReviewMeeting, September 21–23, 2016, Arlington, VA, USA.• DARPA EQUiPS PI ReviewMeeting, March 22–24, 2016, Stanford University, CA, USA.

TEACHING• Introduction to Data Science (using GitHub, R, and Python), University of Colorado Boulder, Spring 2019.• Gaussian Processes andDeep Learning, Tutorial, BrownUniversity, Spring 2017.• Introduction to Linear Algebra, Teaching Assistant, University ofMaryland – College Park, Fall 2015.• Calculus 3, Teaching Assistant, University ofMaryland – College Park, Fall 2014.• Research Experience for Undergraduate Students, GraduateMentor, University ofMaryland – College Park, Summer2014.

• Linear Algebra for Scientists and Engineers, Teaching Assistant, University ofMaryland – College Park, Spring 2014.• Differential Equations, Teaching Assistant, University ofMaryland – College Park, Fall 2013.• Research Experience for Undergraduate Students, GraduateMentor, GeorgeMason University, Summer 2012 & 2013.

SKILLSMatlab ○○○○○Python (NumPy/SciPy) ○○○○○Scikit-learn & StatsModels ○○○○○R ○○○○○Object Oriented Programming ○○○○○C&C++&Cuda ○○○○○Parallel Computing (MPI) ○○○○○Deep Learning ○○○○○Tensorflow ○○○○○PyTorch &Autograd ○○○○○Theano &Caffe ○○○○○Big Data ○○○○○Apache Spark (Scala) ○○○○○Mathematica ○○○○○Fortran ○○○○○Maple ○○○○○SQL ○○○○○Pandas ○○○○○Stata & EViews ○○○○○VBA ○○○○○SAS& SPSS ○○○○○Flask ○○○○○Heroku ○○○○○NLTK& spaCy ○○○○○Bloomberg ○○○○○Numerix ○○○○○Summit ○○○○○

LANGUAGESFarsi (Persian) ○○○○○English ○○○○○French ○○○○○German ○○○○○

REFERENCESGeorge Karniadakis[ [email protected] Division of AppliedMathematics, BrownUniversityStuart Geman[ [email protected] Division of AppliedMathematics, BrownUniversityMichael Triantafyllou[ [email protected] Department ofMechanical Engineering, Massachusetts Institute of TechnologyDilipMadan[ [email protected] Department of Finance, University ofMaryland College ParkPadmanabhan Seshaiyer[ [email protected] Department ofMathematical Sciences, GeorgeMason UniversityPeymanGivi[ [email protected] Department ofMechanical Engineering, University of PittsburghMark Girolami[ [email protected] Department of Statistics, University ofWarwick