Research interests:
My research is in probability theory and its connections to data-science. Some specific areas of interest include
Statistical learning theory with special interest in inference of stochastic dynamical systems, stochastic methods in computing, statistical algorithms
Limit theorems with special interest in large deviation analysis; their applications to statistics and numerical methods
Multiscale stochastic systems, averaging principle and applications to model reduction techniques
Grants:
Travel support for Mathematicians, Simons Foundation, September 1, 2023
NSF DMS - 2246815, August 15, 2023 - July 31, 2026
NSF DMS - 1855788, July 1, 2019- June 30, 2023
LA Board of Regents, July 1, 2016-June 30, 2019
Ph.D. Students:
Current: Majid Abubakar, Maganizo Kapita, Anan Saha
Graduated: Jinpu Zhou (2024). First Position: Research Scientist, Meta
Selected Works:
Nonparametric learning of stochastic differential equations from sparse and noisy data. Co-authors: R. Mitra, J. Zhou. Preprint, 2025. arXiv:2508.11597
Functional limit theorems and parameter inference for multiscale stochastic models of enzyme kinetics. Co-author: W. R. KhudaBukhsh. Preprint, 2025. arXiv:2409.06565
Nonparametric learning of covariate-based Markov jump processes using RKHS techniques. Co-authors: Y. Han and R. Mitra. Preprint, 2025. arXiv:2505.03119
Asymptotic Analysis of the Total Quasi-Steady State Approximation for the Michaelis-Menten Enzyme Kinetic Reactions. Co-author: W. R. KhudaBukhsh. Preprint, 2025. arXiv:2503.20145
Asymptotic analysis of estimators of ergodic stochastic differential equations. Preprint, 2024. arXiv:2411.03623
Optimal learning via moderate deviations theory. Co-author: T. Sutter. Preprint, 2024. arXiv:2305.14496
Moment stability of stochastic processes with applications to control systems. Co-author: D. Chatterjee. Mathematical Control and Related Fields, 14(1), 386-412, 2024. arXiv:2206.00200
Infinite-dimensional optimization and Bayesian nonparametric learning of stochastic differential equations. Co-authors: R. Mitra, J. Zhou. Journal of Machine Learning Research, 24(159),1−39, 2023. arXiv:2205.15368
Inhomogeneous functionals and approximations of invariant distribution of ergodic diffusions: Central limit theorem and moderate deviation asymptotics. Co-author: P. Sundar. Stochastic Processes and their Applications, 133, 74-110, 2021. arXIv:1805.06388
Large Deviations for Small Noise Diffusions in a Fast Markovian Environment. Co-authors: A. Budhiraja, P. Dupuis. Electron. J. Probab., Vol. 23 (2018), paper 112, pp 33. arXiv:1705.0294.
Large deviation principle for stochastic integrals and stochastic differential equations driven by infinite-dimensional semimartingales. Stochastic Processes and their Applications, 128 (7), 2179 - 2227, 2018. arXiv:1101.5515.
A variational approach to path estimation and parameter inference of hidden diffusion processes. Co-authors: T.Sutter, H.Koeppl. Journal of Machine Learning Research, 17(190), 1 - 37, 2016. arXiv:1508.00506.
Moderate Deviation Principles for Stochastic Differential Equations with Jumps. Co-authors: A. Budhiraja, P. Dupuis. Annals of Probability, Vol. 44, No. 3, 1723 -1775, 2016. arXiv:1401.7316.
Efficient simulation of multiscale reaction networks: A multilevel partitioning approach. Co-authors: D. Altintan, H.Koeppl. Proceedings of ACC, 2016, pages 6073 - 6078, Print ISBN: 978-1-4673-8683-8.
Error bound and simulation algorithm for piecewise deterministic approximations of stochastic reaction systems. Co-authors: D. Altintan, H.Koeppl. Proceedings of ACC, 2015, pages 787 - 792, Print ISBN: 978-1-4799-8685-9.
Jump-Diffusion Approximation of Stochastic Reaction Dynamics: Error bounds and Algorithms. Co-authors: D. Altintan, H .Koeppl. SIAM: Multiscale Modeling and Simulation. 13(4), 1390 - 1419, 2015. arXiv:1409.4303.
Markov chain aggregation and its applications to combinatorial reaction networks. Co-authors: T. Petrov, H. Koeppl. Journal of Mathematical Biology, Vol. 69, Issue 3, pp 767-797, 2014. arXiv:1303.4532.
Wong-Zakai type convergence in infinite dimensions. Electron. J. Probab. Vol. 18, No. 31, 34 pp, 2013.
Hybrid spatial Gillespie and particle tracking simulation. Co-authors: M. Klann, H. Koeppl. Bioinformatics, Vol. 28, No. 18, pp. i549-i555, 2012.
Model Decomposition and Stochastic Fragments. Co-authors: T. Petrov, H. Koeppl. Electronic Notes in Theoretical Computer Science, Vol. 284, 105-124, 2012.
Accounting for extrinsic variability in the estimation of stochastic rate constants. Co-authors: H. Koeppl, C. Zechner, S. Pelet, M. Peter. International Journal of Robust and Nonlinear Control, Vol. 22, No. 10, 1103 - 1119, 2012.
Deterministic characterization of phase noise in biomolecular oscillators. Co-authors: H. Koeppl, M. Hafner, A. Ganguly, A. Mehrotra. Physical Biology, Vol. 8, No. 5, 2011.
Error analysis of tau-leap simulation methods. Co-authors: D. F. Anderson, T. G. Kurtz. Annals of Applied Probability, Vol. 21, No. 6, 2226 - 2262, 2011. arXiv:0909.4790v1.