Jessica Cox

CV

Updated

I am a computational statistician, currently working as a postdoctoral researcher at the Max Planck Institute for Physics. My background is in spatio-temporal Bayesian signal processing, in particular particle filters and parameter inference therein. My current work is on model serialisation: software layers for the creation, transmission, evaluation, and long-term storage of statistical models, from which FlatPPL came. Separately, I work on BAT.jl.

Dr Jessica Cox, formerly Benjamin Cox

bcox@mpp.mpg.de · jmcox@posteo.de

Positions

  • Max Planck Institute for Physics, Postdoctoral Researcher, 2026–present

    Model serialisation: developing software layers for the creation, transmission, evaluation, and long-term storage of statistical models, which is where FlatPPL came from. Separately, work on BAT.jl, the Bayesian Analysis Toolkit. Loosely associated with the MADMAX and LEGEND groups.

    My position is funded by Germany’s Federal Ministry of Research, Technology and Space (BMFTR) within the ErUM-Data programme, under grant FKZ 05D25PC1 (DEMOS consortium).

Education

  • University of Edinburgh, PhD Statistics, 2020–2025

    Thesis: Parameter estimation in sparse state-space models (doi:10.7488/era/6823). Supervisor: Víctor Elvira.

  • University of Edinburgh, BSc (Hons) Mathematics, 2016–2020

Awards and grants

  • University of Edinburgh – Rice University Strategic Collaboration Award, 2022
  • NERC SENSE CDT studentship (NE/T00939X/1)

Service

  • Reviewer, IEEE Transactions on Signal Processing

Publications

Preprints

  1. J.-J. Brady, B. Cox, Y. Li and V. Elvira, “PyDPF: A Python Package for Differentiable Particle Filtering,” 2026. arXiv:2510.25693 · package · source · docs

Journal articles

  1. B. Cox, S. Segarra and V. Elvira, “Learning state and proposal dynamics in state-space models using differentiable particle filters and neural networks,” Signal Processing, 2025. doi:10.1016/j.sigpro.2025.109998

  2. B. Cox, É. Chouzenoux and V. Elvira, “GraphGrad: Efficient Estimation of Sparse Polynomial Representations for General State-Space Models,” IEEE Transactions on Signal Processing, vol. 73, pp. 1562–1576, 2025. doi:10.1109/TSP.2025.3554876 · arXiv:2411.15637

  3. B. Cox and V. Elvira, “Sparse Bayesian Estimation of Parameters in Linear-Gaussian State-Space Models,” IEEE Transactions on Signal Processing, vol. 71, pp. 1922–1937, 2023. doi:10.1109/TSP.2023.3278867 · arXiv:2306.11652

Conference papers

  1. B. Cox, É. Chouzenoux and V. Elvira, “Learning a Sparse Polynomial Approximation to the Transition Function of General State-Space Models,” ICASSP 2025, Hyderabad. doi:10.1109/ICASSP49660.2025.10888706

  2. B. Cox, S. Pérez-Vieites, N. Zilberstein, M. Sevilla, S. Segarra and V. Elvira, “End-to-End Learning of Gaussian Mixture Proposals Using Differentiable Particle Filters and Neural Networks,” ICASSP 2024, Seoul, pp. 9701–9705.

  3. M. Sevilla, N. Zilberstein, B. Cox, S. Pérez-Vieites, V. Elvira and S. Segarra, “State and Dynamics Estimation with the Kalman-Langevin Filter,” 57th Asilomar Conference on Signals, Systems, and Computers, 2023, pp. 1372–1376. doi:10.1109/IEEECONF59524.2023.10476814

  4. B. Cox and V. Elvira, “Parameter Estimation in Sparse Linear-Gaussian State-Space Models via Reversible Jump Markov Chain Monte Carlo,” 30th European Signal Processing Conference (EUSIPCO), Belgrade, 2022, pp. 797–801. doi:10.23919/EUSIPCO55093.2022.9909829