Simon Byrne

I am a PhD student in the StatsLab, supervised by Philip Dawid.

My work is focused on graphical models: particularly the parametrisation and specification of prior and posterior laws for the structure of various graphs. A good introduction to my work is this poster I presented at the 9th Valencia meeting on Bayesian statistics.

Papers

• "Graphical models, hyper Markov properties and covariance graphs", Fourth term report, 2009.
• "Message passing inference in graphical models", Part III essay, 2008.

Presentations

• "Priors for DAGs: structural estimation of graphical models", Research Student Conference, Cambridge. 5 April 2011.
• "The structural Markov property: estimating graphical models" (poster), 9th Valencia meeting on Bayesian statistics. 3 June 2010.
• "Graphical models", a general interest talk for the Clare Research Symposium, 11 March 2010.
• "Fourth term review", a summary of the first year and a bit of my PhD, 24 February 2010.
• "Variation independence and case control studies", a 4 minute talk from the Cambridge Statistics Initiative One-Day Meeting, 25 September 2009.
• "Message-passing inference in graphical models", Machine Learning @ CUED seminar, 15 July 2008.