Professor Dino Sejdinovic
Statistical machine learning, reproducing kernel Hilbert spaces, hypothesis testing with big data, measures of association and multivariate interaction, tradeoffs between computational and statistical efficiency, coding and information theory.
Prior to joining the Department in October 2014, I was with the Gatsby Computational Neuroscience Unit, University College London as a postdoctoral fellow in machine learning, and before that held a Brunel fellowship at the Department of Mathematics, University of Bristol. I received a Diplom in Mathematics and Theoretical Computer Science from the University of Sarajevo (2006) and a PhD in Electrical and Electronic Engineering from the University of Bristol (2009).
K. Chwialkowski, D. Sejdinovic and A. Gretton, A wild bootstrap for degenerate kernel tests, Advances in Neural Information Processing Systems (NIPS) 27, Dec. 2014.
D. Sejdinovic, H. Strathmann, M. Lomeli Garcia, C. Andrieu and A. Gretton, Kernel adaptive Metropolis-Hastings, Proc. International Conference on Machine Learning ICML - JMLR W&CP 32(2), p. 1665-1673, 2014.
D. Sejdinovic, A. Gretton and W. Bergsma, A kernel test for three-variable interactions, Advances in Neural Information Processing Systems (NIPS) 26, Dec. 2013.
D. Sejdinovic, B. Sriperumbudur, A. Gretton and K. Fukumizu, Equivalence of distance-based and RKHS-based statistics in hypothesis testing, Annals of Statistics 41(5), p. 2263-2291, Oct. 2013.
O. Johnson, D. Sejdinovic, J. Cruise, A. Ganesh and R. Piechocki, Non-parametric change-point estimation using string matching algorithms, Methodology and Computing in Applied Probability, July 2013.