Kostas Tsampourakis

Kostas Tsampourakis

About me

I recently completed my PhD in Statistics at the School of Mathematics, University of Edinburgh, supervised by Víctor Elvira. Previously I studied physics at the University of Crete and mathematics at Washington State University.

My research interests are in developing robust and efficient Bayesian inference algorithms for complex dynamical systems and signal processing, with a focus on Bayesian filtering and smoothing of state-space models.

News

  • Summer 26 Passed my viva and graduating with a PhD in Statistics from the University of Edinburgh!

  • May 26 Two preprints submitted — one to IEEE Transactions on Signal Processing and one to Statistics and Computing. See below!

  • Nov. 23 Thrilled to accept a poster award at Bayes@CIRM 2023!

  • Oct. 23 Happy to attend and present a poster at the Bayes@CIRM workshop in Marseille, France!

  • Jun. 23 Happy to attend and present a poster at MCM 2023!

  • Jun. 23 Poster presented at IEEE ICASSP 2023 in Rhodes island, Greece.

  • Mar. 23 Glad to attend and present a poster at BayesComp 2023 in Levi, Finland!

  • Mar. 23 Excited to attend the inaugural Probabilistic Numerics Spring School in Tübingen, Germany!

  • Feb. 23 Paper accepted at IEEE ICASSP 2023!

  • May 22 Oral presentation at IEEE ICASSP 2022 in Singapore.

  • Jan. 22 Paper accepted at IEEE ICASSP 2022!

Publications

Preprints

  1. Kostas Tsampourakis, Víctor Elvira. (2025). “A Gaussian Sum Filter for Unifying Gaussian and Particle Filters” Submitted to IEEE Transactions on Signal Processing.
  2. Kostas Tsampourakis, Víctor Elvira. (2025). “Truncated Neural Likelihood Estimation for Simulation-Based Inference in State-Space Models” Submitted to Statistics and Computing.

Peer-reviewed

  1. Kostas Tsampourakis, Víctor Elvira. (2023). “An Augmented Gaussian Sum Filter Through a Mixture Decomposition” 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).
  2. Kostas Tsampourakis, Víctor Elvira. (2022). “Approximating The Likelihood Ratio in Linear-Gaussian State-Space Models for Change Detection” 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).
  3. Abhishek Kaul, Hongjin Zhang, Kostas Tsampourakis, George Michailidis. “Inference on the Change Point under a High Dimensional Covariance Shift” 2023, Journal of Machine Learning Research, vol. 24 .