Location
LGRT 1423G

Office Hours:
M, Th 4-5:00PM in LGRT 1423G or by appointment
  • Markos Katsoulakis is working at the interface of Applied & Computational Mathematics and Data Science, focusing on  machine learning,  uncertainty quantification and applications in  materials and catalysis.
  • He is the Director for the Center of Applied Mathematics at the University of Massachusetts and  a member of the Executive Committee of the  UMASS TRIPODS Institute for Theoretical Foundations of Data Science.  
  • He currently serves in the Editorial Board of the SIAM/ASA Journal of Uncertainty Quantification and the Editorial Board of the SIAM Mathematical Modeling and Computation book series. He was an Associate Editor of SIAM Journal of Mathematical Analysis between 2002-2014.
  • Brief CV (2024)

Education

Ph.D. Brown University, 1993

B.A. National and Kapodistrian University, 1987

RESEARCH INTERESTS

Generative Modeling, Scientific Machine Learning, Information Theory, Uncertainty Quantification, Multi-scale Methods

Teaching

Math 690STF 2024: Mathematics of Generative Artificial Intelligence

In the last decade, generative models and generative artificial intelligence have produced breakthrough results in image generation, text and speech synthesis, and more recently in scientific research itself, in fields such as aerospace, astronomy, biology, materials science, and medicine. Not unexpectedly, mathematics plays a foundational role in generative AI. It is essential for gaining a deeper understanding of existing methods, establishing limitations, quantifying trustworthiness, and developing new, provably more robust, or more energy-efficient methods.  We will also discuss applications of generative methods in Applied Sciences and Engineering that require a combination of GenAI with physical models (PDEs, dynamical systems) such as: likelihood-free inference and inverse problems, foundation models for PDEs, solving very high-dimensional control problems (particle systems, swarms, etc). Students can access the class, a detailed syllabus, slides for all lectures and other course material through Canvas

Selected Publications

  1. N. Mimikos-Stamatopoulos, B. J. Zhang, M. A. Katsoulakis, Score-based generative models are provably robust: an Uncertainty Quantification perspectiveThirty-Eighth Annual Conference on Neural Information Processing Systems, NeurIPS 2024
  2. H. Gu, P. Birmpa, Y. Pantazis, M. A. Katsoulakis, and L. Rey-Bellet, Lipschitz-regularized gradient flows and generative particle algorithms for high-dimensional scarce data, SIAM Data Science, to appear, (2024).
  3. B. J. Zhang, M. A. Katsoulakis, A mean-field games laboratory for generative modeling, preprint
  4. Z. Chen, M. A. Katsoulakis, L. Rey-Bellet, W.  Zhu, Sample Complexity of Probability Divergences under Group Symmetry)International Conference on Machine Learning ICML 2023 
  5. J. Birrell, P. Dupuis, M. A. Katsoulakis, Y. Pantazis, L. Rey-Bellet, Function-space regularized Rényi divergencesInternational Conference on Learning Representations ICLR 2023
  6. J. Birrell, P. Dupuis, M. A. Katsoulakis, Y. Pantazis, L. Rey-Bellet, (f, Γ) -Divergences: Interpolating between f-Divergences and Integral Probability MetricsThirty-sixth Conference on Neural Information Processing Systems NeurIPS 2022 
  7. J. Birrell,  M. A. Katsoulakis, L. Rey-Bellet, W. Zhu, Structure-preserving GANs39th International Conference on Machine Learning ICML 2022  
  8. P. Birmpa, J. Feng, M. A. Katsoulakis, L. Rey-Bellet, Model Uncertainty and Correctability for Directed Graphical Models, SIAM/ASA Journal on Uncertainty Quantification 10 (4),1461-1512, (2022)
  9. Y. Pantazis, D. Paul, M. Fasoulakis, Y. Stylianou, M. A. Katsoulakis, Cumulant GANIEEE Transactions on Neural Networks and Learning Systems, (2022) 
  10. J. Birrell, M. A. Katsoulakis, Y. Pantazis, Optimizing variational representations of divergences and accelerating their statistical estimation, IEEE Transactions on Information Theory, (2022) 
  11. J. Birrell, P. Dupuis, M. A. Katsoulakis, Y. Pantazis, L. Rey-Bellet, (f, Γ) -Divergences: Interpolating between f-Divergences and Integral Probability Metrics,  Journal of Machine Learning Research, 23 (39), 1-70, (2022).  
  12. P. Birmpa, M. A. Katsoulakis, Uncertainty Quantification for Markov Random Fields SIAM/ASA J. Uncertainty Quantification, 9(4), 1457–1498, (2021).
  13. Eric J. Hall, Søren Taverniers, Markos A. Katsoulakis, Daniel M. Tartakovsky, GINNs: Graph Informed Neural Networks for multiscale physicsJ. Comp. Phys. Volume 433, 110192 (2021).
  14. Feng J, Lansford J, Katsoulakis M, Vlachos D. Explainable and trustworthy artificial intelligence for correctable modeling in chemical sciences, Science Advances. 2020 October 14; 6(42):eabc3204
  15. K. Um, E. J. Hall, M. A. Katsoulakis, D. M. Tartakovsky. Causality and Bayesian Network PDEs for multiscale representations of porous media, J. Comp. Phys. 394:658-678 (2019).
  16. P. Dupuis, M. A. Katsoulakis, Y. Pantazis, L. Rey-Bellet Sensitivity Analysis for Rare Events based on Renyi Divergence, Annals Applied Prob., 2020 August; 30(4):1507-1533.
  17. Gourgoulias K, Katsoulakis M, Rey-Bellet L, Wang J. How Biased Is Your Model? Concentration Inequalities, Information and Model Bias, IEEE Transactions on Information Theory. 2020; 66(5):3079-3097.
  18. P. Vilanova, M. A. Katsoulakis, Data-driven, variational model reduction of high-dimensional reaction networksJ. Comp. Phys. Volume 401, 108997 (2020)
  19. V. Harmandaris, E. Kalligiannaki, M. A. Katsoulakis, P. Plechac. Path-space Variational Inference for non-equilibrium coarse-grained systems, J. Comp. Phys. 314,355-383 (2016).
  20. P. Dupuis, M. A. Katsoulakis, Y. Pantazis, P. Plechac. Path-Space Information Bounds for Uncertainty Quantification and Sensitivity Analysis of Stochastic Dynamics, SIAM/ASA J. Uncertainty Quantification, 4(1), 80-111, (2016).
  21. J. E. Sutton, W. Guo, M. A. Katsoulakis & D. G. Vlachos. Effects of correlated parameters and uncertainty in electronic-structure-based chemical kinetic modelling, Nature Chemistry 8, 331–337 (2016).
  22. G. Arampatzis, M. A. Katsoulakis and P. Plechac,  Parallelization, processor communication and error analysis in lattice kinetic Monte CarloSIAM Num. Analysis, 52, no. 3, 11561182, (2014).
  23. Y. Pantazis, D.G. Vlachos and M. A. Katsoulakis, Parametric Sensitivity Analysis for Biochemical Reaction Networks based on Pathwise Information Theory, BMC Bionformatics 14:311, (2013).
  24. G. Arampatzis, M. A. Katsoulakis, P. Plechác, M. Taufer, L. Xu. Hierarchical fractional step approximations and parallel kinetic Monte Carlo algorithms. J. Comp. Phys. 231, 7795-7814, (2012).
  25. J. Feng ,  M. A. Katsoulakis, A Hamilton Jacobi theory for controlled gradient flows in infinite dimensionsArch. Rat. Mech. Analysis 192, 2 , 275-310 (2009).
  26. A. Sopasakis & M. A. Katsoulakis, Stochastic modeling and simulation of traffic flow: ASEP with Arrhenius look-ahead dynamicsSIAM J. Appl. Math., 66, 921-944, (2006).
  27. T. M. Davis, T. O.Drews, H. Ramanan, C. He, J. Dong, H. Schnablegger, M. A. Katsoulakis, E. Kokkoli, A. V. McCormick, R. L.Penn, and M. Tsapatsis. Mechanistic Principles of Nanoparticle Evolution to Zeolite CrystalsNature Materials, 5, 400, (2006)
  28. A. Chaterjee, M. A. Katsoulakis & D. G. Vlachos, Binomial distribution based τ -leap accelerated stochastic simulationJ. Chem. Phys. 122, 024112 (2005)
  29. M. A. Katsoulakis, A. J. Majda, D. G. Vlachos. Coarse-grained stochastic processes for lattice systemsProc. Natl. Acad. Sci. USA 100, 782-787, (2003).
  30. M. A. Katsoulakis & A.E. Tzavaras, Contractive Relaxation Systems and the Scalar Multidimensional Conservation LawComm. P.D.E. 22, (1997), 195-233.
  31. M. A. Katsoulakis & P. E. Souganidis, Generalized motion by mean curvature as a macroscopic limit of stochastic Ising models with long range interactions and Glauber dynamicsComm. Math. Phys., 169, (1995), 61-97.

For a complete list see:  Google Scholar page & Researchgate profile