Academic

Hello, my name is Marcelo Pereyra and I am a Research Fellow in the Probability and Statistics group of the School of Mathematics of the University of Bristol, where I hold a Marie Curie Intra-European Fellowship for Career Development. I am also Associate Editor for Digital Signal Processing, and Guest Editor for the special issue on "Stochastic Simulation and Optimisation in Signal Processing" of the IEEE Journal of Selected Topics in Signal Processing..

Research

My research activities lie at the intersection of image processing, computational statistics, and optimisation. Most of my recent work is on Bayesian analysis and computation methods for high-dimensional inverse problems related to mathematical imaging. I am particularly interested in new stochastic simulation and optimisation approaches and in connections between MCMC sampling and convex analysis and optimisation (e.g. Proximal MCMC algorithms).

News 


     New     
NEWs
New paper "Maximum-a-posteriori estimation with Bayesian confidence regions", submitted to SIAM Journal on Imaging Sciences. E-print on Arxiv
     New     
NEWs
I will be giving a tutorial on "Stochastic Simulation and Optimisation in Image Processing" at the Peyresq 2016 summer school of the French signal processing society
Jan. 2016
NEWsNEWs
I am organising a SuSTaIn EdgeCutter workshop on "High-dimensional Statistics, Inverse Problems and Convex Analysis" with Dr. Carola Schönlieb. The workshop will take place at the Royal Statistical Society, London, UK, on 22nd March 2016. Please see this web page for more details (notice the fantastic speakers!).
Dec. 2015
NEWsNEWs
New paper "Fast unsupervised Bayesian image segmentation with adaptive spatial regularisation" with S. McLaughlin, submitted to IEEE Trans. Image Processing. E-print on Arxiv
Nov. 2015
NEWsNEWs
New paper "A Survey of Stochastic Simulation and Optimization Methods in Signal Processing" with P. Schniter, E. Chouzenoux, J.-C. Pesquet, J.-Y. Tourneret, A. Hero, and S. McLaughlin, IEEE Sel. Topics in Signal Processing, in press. E-print on Arxiv
Oct. 2015
NEWsNEWs
New paper "Bayesian nonlinear hyperspectral unmixing with spatial residual component analysis," with Y. Altmann and S. McLaughlin, IEEE Trans. Image Processing, in press. E-print on Arxiv
Oct. 2015
NEWsNEWs
New paper "Collaborative sparse regression using spatially correlated supports - Application to hyperspectral unmixing," with Y. Altmann and J. Bioucas-Dias, IEEE Trans. Image Processing, in press. E-print on Arxiv
Jul. 2015
NEWsNews
New paper "Bayesian computation: a perspective on the current state, and sampling backwards and forwards", with P. Green, K. Łatuszynski, and C. P. Robert, Statistics and Computing, vol. 25, no. 4, pp 835-862, Jul. 2015. Arxiv
May 2015  New paper "Proximal Markov chain Monte Carlo algorithms", Statistics and Computing. (open access paper)
Mar. 2015
NEWsNEWs
I'm a guest editor for the special issue on "Stochastic Simulation and Optimisation in Signal Processing", IEEE Journal of Selected Topics in Signal Processing. Please see the Call for Papers for details (CFP closed on 14/05/15).
Feb. 2015
NEWsNEWs
New paper "Exploiting information geometry to improve the convergence properties of nonparametric active contours" with H. Batatia and S. McLaughlin, IEEE Trans. Image Processing, vol. 24, no. 3, Mar. 2015.
Jan. 2015
NEWsNEWs
I have started a Marie Curie Intra-European Fellowship for Career Development entitled "Stochastic Optimisation and Simulation for Image Processing" with Prof. Christophe Andrieu, School of Mathematics, University of Bristol.


Short Bioschetch

Marcelo Pereyra was born in Buenos Aires, Argentina, in 1984. He studied electronic engineering and received a double M.Eng. degree from ITBA (Argentina) and INSA Toulouse (France), together with a M.Sc. degree from INSA Toulouse, in June 2009. In July 2012 he obtained a Ph.D. degree from University of Toulouse. He currently holds a Marie Curie Intra-European Fellowship for Career Development at the School of Mathematics of the University of Bristol. He is also the recipient of a Brunel Postdoctoral Research Fellowship in Statistics, a Postdoctoral Research Fellowship from French Ministry of Defence, a Leopold Escande PhD Thesis excellent award from the University of Toulouse (2012), an INFOTEL R&D excellent award from the Association of Engineers of INSA Toulouse (2009), and an ITBA R&D excellence award from the Buenos Aires Institute of Technology (2007).

Contact

Department of Mathematics
University of Bristol
University Walk, Clifton
Bristol BS8 1TW, U.K.
Office : 3.13 - 3rd floor
Tel: +44 (0) 117 928 7999
E-mail: marcelo [dot] pereyra [at] bristol [dot] ac [dot] uk
www: https://people.maths.bris.ac.uk/~mp12320/