Signal Processing and Communications Laboratory

Department of Engineering

Ioannis Papageorgiou Ioannis Papageorgiou

Background - Research - Publications

Position: PhD Student

E-mail: ip307 [at] cam.ac.uk

Office Location: BN3-02

Thesis Title: Time series modelling and inference using Bayesian Context Trees

Supervisor: Ioannis Kontoyiannis

Background

I obtained my BA and MEng degrees at the University of Cambridge, with Distinction in Information and Computer Engineering. My MEng thesis was titled "Active Reinforcement Learning", and it was on a model-based Reinforcement Learning approach that uses Gaussian Processes. Here is a link to my CV.

Research Interests

My research interests lie at the interface of Bayesian Statistics, Machine Learning, and Information Theory. I am particularly interested in combining modern machine learning techniques with ideas from information theory, which come with strong theoretical foundations and guarantees. My primary focus has been on the modelling and inference of time series (both discrete-valued and real-valued), and more specifically on developing Bayesian methods that are inspired by information-theoretic ideas and algorithms, such as context-tree weighting. Here is a link to my Google Scholar page.

Publications

Journal papers

  • I. Kontoyiannis, L. Mertzanis, A. Panotopoulou, I. Papageorgiou, and M. Skoularidou. Bayesian Context Trees: Modelling and exact inference for discrete time series. Journal of the Royal Statistical Society: Series B (Statistical Methodology) , 84(4):1287–1323, 2022.

Preprints

  • I. Papageorgiou and I. Kontoyiannis. The Bayesian Context Trees State Space Model: Interpretable mixture models for time series. arXiv preprint arXiv:2106.03023, submitted for publication, 2022.
  • I. Papageorgiou and I. Kontoyiannis. Posterior representations for Bayesian Context Trees: Sampling, estimation and convergence. arXiv preprint arXiv:2202.02239, submitted for publication, under minor revision in Bayesian Analysis, 2022.
  • V. Lungu, I. Papageorgiou, and I. Kontoyiannis. Change-point Detection and Segmentation of Discrete Data using Bayesian Context Trees. arXiv preprint arXiv:2203.04341, submitted for publication, 2022.

Conferences (papers/talks)

  • I. Papageorgiou and I. Kontoyiannis. The Posterior Distribution of Bayesian Context-Tree Models: Theory and Applications. In 2022 IEEE International Symposium on Information Theory (ISIT), pp. 702–707, 2022.
  • I. Papageorgiou, I. Kontoyiannis, L. Mertzanis, A. Panotopoulou, and M. Skoularidou. Revisiting context-tree weighting for Bayesian inference. In 2021 IEEE International Symposium on Information Theory (ISIT), pp. 2906–2911, 2021.
  • V. Lungu, I. Papageorgiou, and I. Kontoyiannis. Bayesian Change-Point Detection via Context-Tree Weighting. In 2022 IEEE Information Theory Workshop (ITW), 2022.
  • "Modelling and inference for time series using Bayesian Context Trees". In Greek Stochastics ′, Corfu, Greece, August 2022.
  • "Bayesian mixture models for time series based on context trees". In 36th International Workshop on Statistical Modelling (IWSM), Trieste, Italy, July 2022.
  • "Bayesian autoregressive mixture models based on context trees". In 42nd International Symposium on Forecasting (ISF), Oxford, UK, July 2022.
  • "Modelling and inference for discrete time-series using Bayesian Context Trees". In IMS Annual Meeting in Probability and Statistics, London, UK, June 2022.

Software

  • I. Papageorgiou, V. Lungu, and I. Kontoyiannis. R package BCT: "Bayesian Context Trees for Discrete Time Series". Available at: https://CRAN.R-project.org/package=BCT, version 1.1, November 2020; version 1.2, May 2022.