mattstats.wordpress.com
Thesis submitted, articles in press | Matt Moores
https://mattstats.wordpress.com/2015/02/21/thesis-submitted-articles-in-press
Thesis submitted, articles in press. February 21, 2015. Some of my thesis papers have recently appeared online:. 8220;Pre-processing for approximate Bayesian computation in image analysis”. Moores, Drovandi, Mengersen and Robert (2015). 1): 23-33. DOI: 10.1007/s11222-014-9525-6. 8220;An external field prior for the hidden Potts model, with application to cone-beam computed tomography”. Moores, Hargrave, Deegan, Poulsen, Harden and Mengersen (2015). Computational Statistics and Data Analysis. July 12, 2016.
mattstats.wordpress.com
Graduation Ceremony | Matt Moores
https://mattstats.wordpress.com/2015/07/29/graduation-ceremony
July 29, 2015. I was very glad that I was able to attend my PhD graduation ceremony in Brisbane last week. My extended family were there to cheer me on, as well as both of my supervisors and my dear friend, Kal. Out of 28 doctorates there were a dozen from maths, including four statisticians. Unfortunately none of us snagged an Outstanding Doctoral Thesis Award, but it was nice to be nominated! Brisbane QLD, Australia. From → Uncategorized. Leave a Reply Cancel reply. Enter your comment here. June 3, 2016.
mattstats.wordpress.com
Talk at QUT on Friday August 7 | Matt Moores
https://mattstats.wordpress.com/2015/08/06/talk-at-qut-on-friday-august-7
Talk at QUT on Friday August 7. August 6, 2015. Lorentzian mixture model for Raman spectroscopy. Matt Moores (University of Warwick, UK). Friday 7 August 2015. 3:00pm – 4:00pm. O603, Gardens Point Campus, Queensland University of Technology. This is joint work with Mark Girolami. Warwick), Kirsten Gracie. Brisbane QLD, Australia. From → Functional Data. Talk at Oxford on Friday March 11 Matt Moores. Bayesian modelling and quantification of Raman spectroscopy Matt Moores. Leave a Reply Cancel reply. Ingma...
shogun-toolbox.org
The SHOGUN Machine Learning Toolbox
http://www.shogun-toolbox.org/page/planet/shogun
You are here: planet/shogun. March 01, 2016 11:15 AM. Google Summer of Code 2016. Just got accepted to the GSoC 2016. After our break year. In 2015, we are extremely excited to continue our GSoC tradition. When I first joined Shogun). If you are a student and wish to spend the summer hacking Machine Learning, guided by a vibrant international community of academics, professionals, and NERDS, then pay us a visit. Oh, and you will receive a cheque over $5000 from Google. Check our our ideas list. Of kernel...
mattstats.wordpress.com
Big Data, Big Models, it is a Big Deal | Matt Moores
https://mattstats.wordpress.com/2014/08/13/big-data-big-models-it-is-a-big-deal
Big Data, Big Models, it is a Big Deal. August 13, 2014. I’ve submitted a poster to the Network on Computational Statistics and Machine Learning (NCSML) workshop “Big Data, Big Models, it is a Big Deal” at the University of Warwick on the 1st and 2nd of September. More details are available from the workshop homepage. My abstract is as follows:. Scalable Bayesian computation for intractable likelihoods in image analysis. From → MCMC. Leave a Reply Cancel reply. Enter your comment here. November 4, 2016.
mattstats.wordpress.com
International Workshop on Monte Carlo Methods for Spatial Stochastic Systems | Matt Moores
https://mattstats.wordpress.com/2015/06/24/international-workshop-on-monte-carlo-methods-for-spatial-stochastic-systems
International Workshop on Monte Carlo Methods for Spatial Stochastic Systems. June 24, 2015. I will be presenting a talk at the ACEMS International Workshop on Monte Carlo Methods for Spatial Stochastic Systems. MCMSS) at the University of Queensland, Brisbane, July 21-23 (abstract below). Other speakers include Gareth Roberts. Is now available online. I’ll also be giving a practice talk at the Warwick Young Researchers’ Meeting. Scalable Inference for the Inverse Temperature of a Hidden Potts Model.
mattstats.wordpress.com
R package bayesImageS | Matt Moores
https://mattstats.wordpress.com/2015/03/04/r-package-bayesimages
March 4, 2015. My R package, bayesImageS. Version 0.1-21, is now available online. It uses a hidden Potts model with additive Gaussian noise for image segmentation of 2D and 3D datasets. The latent labels z. Can be simulated using chequerboard updating or the Swendsen-Wang algorithm. Several methods for full Bayesian inference with intractable likelihoods are supported, including pseudolikelihood, the exchange algorithm, path sampling, and approximate Bayesian computation (ABC-MCMC and ABC-SMC). June 3, ...
herrstrathmann.de
Adaptive Kernel Sequential Monte Carlo | herr strathmann
http://herrstrathmann.de/adaptive-kernel-sequential-monte-carlo
Coffee climbing jazz math. Adaptive Kernel Sequential Monte Carlo. December 24, 2015. December 28, 2015. And when your name is Dino Sejdinovic, you can actually integrate out all steps in feature space…. Wrote a nice blog post. About our recently arxived paper draft. It is about using the Kameleon Kernel Adaptive Metropolis-Hastings. Proposal as an MCMC rejuvenation step in a Sequential Monte Carlo. Google Summer of Code 2016. Leave a Reply Cancel reply. Your email address will not be published.
mattstats.wordpress.com
bayesImageS::swNoData(..) | Matt Moores
https://mattstats.wordpress.com/2013/12/13/swnodata
December 13, 2013. This is the first in a series of posts describing the functions and algorithms that I have implemented in the. Gibbs sampling was originally designed by Geman and Geman (1984) for drawing updates from the Gibbs distribution, hence the name. However, single-site Gibbs sampling exhibits poor mixing due to the posterior correlation between the pixel labels. Thus it is very slow to converge when the correlation (controlled by the inverse temperature β) is high. Function in the PottsUtils.
