Sahani Pathiraja

Postdoctoral Researcher

About Me

I am a postdoctoral researcher at the Institute of Mathematics, University of Potsdam, Germany in Prof. Sebastian Reich's group.  I received degrees in both mathematics and engineering and am a University Medallist.  

​

One of my overarching goals is to develop stronger connections between the mathematical & statistical foundations of data science methods and their applications.  I am motivated by 1) how applications can inspire new theory and 2) how theory be developed in a more practically relevant way.

------------------------------ 

​

I am currently a researcher in the Collaborative Research Centre on Data Assimilation SFB1294 at the University of Potsdam in Project A02:Long time stability and accuracy of ensemble transform filters.  Here I mainly focus on studying the theoretical properties of a new class of methods for sequential Bayesian filtering in the non-linear, non-Gaussian setting. These methods involve interacting particle systems characterised by a controlled stochastic differential equation in the mean field limit. 

​

News

• I was recently accepted to the FRIAS Young Academy for Sustainability Research.  Looking forward to our first meeting in October! 
   

• I am organising a mini-symposium at SIAMUQ22 with Hakon Hoel called Advanced methods for high dimensional Bayesian inference and nonlinear filtering.

Research Interests

Analysis 

• stochastic analysis of controlled particle filters, incl. the feedback particle filter and its diffusion map based approximation  

• localisation methods in ensemble Kalman filtering

• interacting particle systems

Algorithms

Applications

• computational Bayesian inference 

• model uncertainty quantification

• time varying parameter estimation based on sequential filtering 

• computational approaches to optimal transport based ensemble smoothing (e.g. second order corrections)

• hydrology, rainfall-runoff modelling

• intracranial hemodynamics 

• mathematics for sustainability science 

Publications

​Stochastic analysis of data science methods  

Algorithmic & Methodological developments

Applications - Hydrology & Hemodynamics

​

1. Pathiraja*, S., Stannat, W. (2021) Analysis of the feedback particle filter with diffusion map based approximation of the gain, Foundations of Data Science, 3(3), pp.615-645. doi:10.3934/fods.2021023  arXiv:2109.02761
    

2. Pathiraja*, S., Reich, S., Stannat, W. (2021) McKean-Vlasov SDEs in non-linear filtering, SIAM Journal on Control and Optimization, [accepted]. arXiv:2007.12658
    

3. Pathiraja*, S. (2020) L^2 convergence of smooth approximations of stochastic differential equations with unbounded coefficients, Stochastic Analysis & Applications, [under review]. arXiv:2011.13009.
    

4. Bishop, A. N., del Moral, P. and Pathiraja, S. (2018) Perturbations and projections of Kalman-Bucy semigroups, Stochastic Processes and their Applications, 128(9). doi:10.1016/j.spa.2017.10.006. arXiv:1701.05978
   
   
   
   
    

5. Pathiraja*, S., van Leeuwen, P. (2021) Multiplicative non-Gaussian model error estimation in data assimilation, Journal of Advances in Modeling Earth Systems, [under review]. arXiv:1807.09621
    

6. de Wiljes, J., Pathiraja, S. and Reich, S. (2020) Ensemble transform algorithms for nonlinear smoothing problems, SIAM Journal on Scientific Computing, 42(1), pp.A87-A114. doi: 10.1137/19M1239544. arXiv:1901.06300
    

7. Pathiraja, S. and Reich, S. (2019). Discrete gradients for computational Bayesian inference. Journal of Computational Dynamics, 6(2), pp.385-400. doi:10.3934/jcd.2019019. arXiv:1903.00186
    

8. Pathiraja*, S., Moradkhani, H., Marshall, L., Sharma, A. and Geenens, G. (2018) Data-driven model uncertainty estimation in hydrologic data assimilation, Water Resources Research, 54(2), pp. 1252-1280. doi: 10.1002/2018WR022627.
    

9. Pathiraja*, S., Anghileri, D., Burlando, P., Sharma, A., Marshall, L. and Moradkhani, H. (2018) Time-varying parameter models for catchments with land use change: the importance of model structure, Hydrology and Earth System Sciences, 22(5), pp. 2903-2919. doi: 10.5194/hess-22-2903-2018.
    

10. Moradkhani, H., Nearing, G., Abbaszadeh, P. and Pathiraja, S. (2018) Fundamentals of data assimilation and theoretical advances, in Duan, Q., Pappenberger, F., Thielen, J., Wood, A., Cloke, H. L., and Schaake, J. C. (eds) Handbook of Hydrometeorological Ensemble Forecasting. Berlin, Heidelberg: Springer Berlin Heidelberg, pp. 1-26. doi:10.1007/978-3-642-40457-3 30-1.
     

11. Pathiraja*, S., Marshall, L., Sharma, A. and Moradkhani, H. (2016) Hydrologic modeling in dynamic catchments: a data assimilation approach, Water Resources Research, 52, pp. 3350-3372. doi: 10.1002/2015WR017192. 
    
    
    
    
     

12. Gaidzik, F., Pathiraja, S., Saalfeld, S., Stucht, S., Speck, O., Thevenin, D., Janiga, G. (2020) Hemodynamic data assimilation in a subject-specific Circle of Willis geometry. Clinical Neuroradiology, doi: 10.1007/s00062-020-00959-2.
     

13. Pathiraja*, S., Anghileri, D., Burlando, P., Sharma, A., Marshall, L. and Moradkhani,H. (2018) Insights on the impact of systematic model errors on data assimilation performance in changing catchments, Advances in Water Resources. 113(December 2017), pp. 202-222. doi:10.1016/j.advwatres.2017.12.006.
     

14. Pathiraja*, S., Marshall, L., Sharma, A. and Moradkhani, H. (2016) Detecting non-stationary hydrologic model parameters in a paired catchment system using data assimilation, Advances in Water Resources. 94, pp. 103-119. doi: 10.1016/j.advwatres.2016.04.021.
     

15. Pathiraja, S.,Westra, S. and Sharma, A. (2012) Why continuous simulation? The role of antecedent moisture in design flood estimation, Water Resources Research. 48(6). doi: 10.1029/2011WR010997.

*= articles where I am the corresponding author

Current Projects

• Stochastic analysis of the diffusion map approximation of the Feedback Particle Filter with Wilhelm Stannat (TU Berlin), Sebastian Reich & Jana de Wiljes (Uni Potsdam) 
   

• Generalisations of McKean-Vlasov SDEs and their properties for nonlinear filtering with Wilhelm Stannat (TU Berlin), Sebastian Reich & Jana de Wiljes (Uni Potsdam) 
   

• Uncertainty quantification of neural network based methods for medical image segmentation with Franziska Gaidzik, Soumick Chatterjee & Gabor Janiga (OvGU Magdeburg)
   

• Ensemble filtering and Bayesian inverse problems with Jana de Wiljes (Uni Potsdam), Lassi Roininen & Heikki Haario (LUT)​

Recent & Upcoming Talks

• Lappeenranta University of Technology Seminar, Finland (August 2021)
   

• SIAM Conference on Computational Science and Engineering, Texas, USA (March 2021)
   

• Data Assimilation Research Centre (DARC) Seminar, University of Reading, United Kingdom (March 2021) 

CV

For further details see my CV

Contact Information

Institute of Mathematics
University of Potsdam
house 29, room 2.52
Karl-Liebknecht Str. 24-25 
14476 Potsdam-Golm 
Germany 

​

tel: +49 331 977 203158 

email: pathiraja (at) uni-potsdam (dot) de