Modelling changepoints in a Bayesian way is elegant and computationally efficient. I am currently working to extend this into a spatio-temporal context and to enable scalable robust inference on multivariate data.
- Knoblauch, J. & Damoulas T. (2018). Spatio-temporal Bayesian On-line Changepoint Detection with Model Selection, International Conference on Machine Learning (ICML-18), to appear.
- Knoblauch, J., Jewson, J. & Damoulas T. (2018). Doubly Robust Bayesian Inference for Non-Stationary Streaming Data using β-Divergences (work in progress)
Recent & Upcoming Talks
- Bayesian On-line Changepoint Detection and Model Selection in high-dimensional data, Workshop on Computational Strategies for Large-Scale Statistical Data Analysis by the International Centre for Mathematical Sciences, Edinburgh (02/07/2018-06/07/2018)
- Extending Bayesian On-line Changepoint Detection, Seminar Series of the CDT in Data Science, University of Edinburgh, 04/07/2018
- Spatio-temporal Bayesian On-line Changepoint Detection with Model Selection, Statistics Seminar of the Glasgow School of Mathematics & Statistics, University of Glasgow, 06/07/2018
Recent & Upcoming Talks by Collaborators
My friend and collaborator Jack Jewson will be presenting novel robust Bayesian techniques for streaming data which we recently started developing together:
- World Meeting of the International Society for Bayesian Analysis (ISBA), Edinburgh (25/06/2018-29/06/2018) [poster presentation]
- Bayesian Young Statisticians Meeting (BAYSM), University of Warwick (02/07/2018-03/07/2018)
My supervisor Theo Damoulas will also be talking about our work at the Alan Turing Institute in two talks for their industrial partners (Accenture & Wiselinx), 21/05/2018-26/05/2018
I have written a substantial amount of software in Python accompanying my research, through which we have been nominated as Turing Reproducible Research Champions 2018 by the Alan Turing Institute. As soon as it is failsafe, the software will be made publicly available.
Office: C1.02, Zeeman building