The paper Prequential posteriors Shreya Roy wrote as part of her PhD thesis at Warwick U, under the supervision of Rito Dutta, Richard Everitt and myself, got published on-line after earlier acceptance by the Japanese Journal of Statistics and Data Science, an official journal of the Japanese Federation of Statistical Science Associations. Coïncidental but unrelated to my Akaike Memorial Lecture prize. The paper will be part of a special issue on Recent Advances in Dynamical Monte Carlo Methods. Congrats to Shreya, soon to defend her viva in Warwick!
Archive for defense
prequential posteriors in the Japanese Journal of Statistics and Data Science
Posted in Books, Statistics, Travel, University life with tags Akaike Lecture, data assimilation, deep generative forecasting models, defense, generalised Bayesian inference, generative models, intractable likelihood, Japan, Japanese Federation of Statistical Science Associations, Japanese Journal of Statistics and Data Science, Monte Carlo methods, prequential loss, prequential priors, publication, sequential Monte Carlo, special issue, viva on September 22, 2026 by xi'anno thesis no more?!
Posted in Kids, University life with tags defense, Julian Besag, Nature, PhD thesis on August 5, 2016 by xi'an
“The traditional goal is to demonstrate the candidate’s ability to conduct independent research on a novel concept and to communicate the results in an accessible way. Where the academics differ is on how best to achieve that goal.”
Nature had an editorial and more on the changing nature of the PhD thesis and the possible abandonment of the thing. This is an interesting if radical proposal. There are many cases of highly successful research careers that bypassed the PhD station. Take for instance Julian Besag. On the one hand, what matters most for granting a PhD is the ability to produce independent research that is innovative enough. For this purpose, publication in a serious scientific journal is the right filter. (Serious as opposed to predatory.) Asking referees to review a thesis when the chapters have been published and hence refereed sounds like a waste of time. Nature also mentions the issue of the oral defence, which varies a lot across countries and institutions from inexistent to highly formal. If the relevance of the oral defence is to assess the ability of the candidate to present one’s work in an understandable manner, conferences should do. Except that talks are never assessed. If a speaker is poor, he or she may not get invited again by those who attended the talk. May. But this is somewhat secondary in that examples abound of geniuses who were or are unable to deliver good lectures, with no consequence on the quality of their research and collaborations. Collaborations is actually a sensitive aspect, as more and more papers that make the PhD thesis are written jointly. Evaluations of the contribution of the PhD candidate then get delicate, especially when several PhDs are involved. (I used to refrain from co-signing publications with my students during their thesis, but I have loosened this rule in the past years as I find myself more involved in some projects and hence more eager or impatient to see the outcome completed!)
On the other hand, a PhD thesis may help in getting students to focus on broader issues, when compared with published short papers on marginal improvements in quick succession. But this may not be enough of an incentive. The status of the PhD student is also somewhat unique and provides a buffer between studies and research position, where the student gradually morphs into a researcher (or gives up). If we were to abandon the PhD thesis, there would need to be some equivalent structure to give them status and financial support. However, most places accommodate graduate researchers and the ability to support them for variable periods. It would just mean adjusting for longer durations and some degree of protection…
[The above picture is copied from the site theses.fr that compiles theses published in France and produces some basic statistics, which are all wrong in my case!]
computer strategies for complex Bayesian models
Posted in Books, Kids, Statistics, University life with tags defense, delayed acceptance, dimension expansion, Gaussian copula, Paris, PhD students, PhD thesis, Porte Dauphine, Université Paris Dauphine on July 18, 2016 by xi'an
This is the cover page of Marco Banterle‘s thesis, who will defend on Thursday [July 21, 13:00], at a rather quiet time for French universities, which is one reason for advertising it here. The thesis is built around several of Marco’s papers, like delayed acceptance, dimension expansion, and Gaussian copula for graphical models. The defence is open to everyone, so feel free to join if near Paris-Dauphine!
Rémi Bardenet’s seminar
Posted in Kids, pictures, Statistics, Travel, University life with tags ABC in Roma, big data, BiPS, CREST, defense, ENSAE, Institut Henri Poincaré, MCMC algorithms, Monte Carlo Statistical Methods, Nicolas Chopin, PhD thesis, quasi-Monte Carlo methods, seminar, tall data on April 7, 2016 by xi'an
Next week, Rémi Bardenet is giving a seminar in Paris, Thursday April 14, 2pm, in ENSAE [room 15] on MCMC methods for tall data. Unfortunately, I will miss this opportunity to discuss with Rémi as I will be heading to La Sapienza, Roma, for Clara Grazian‘s PhD defence the next day. And on Monday afternoon, April 11, Nicolas Chopin will give a talk on quasi-Monte Carlo for sequential problems at Institut Henri Poincaré.
day of the theses
Posted in Statistics, University life with tags defense, habilitation, PhD thesis, Supéléc, Université Paris Dauphine on May 14, 2012 by xi'an
Today, I will spend my day in thesis defenses, as I take part in a defense committee this morning at Supéléc, about a thesis written by Alireza Roodaki on a new approach to trans-dimensional MCMC for mixtures of distributions. Rather than a new way to simulate from posterior distributions with a varying number of components, the thesis concentrates on the post-simulation processing of the outcome of the simulation, constructing an object similar to the point process representation of Matthew Stephens where components have a meaning across varying dimensions. An interesting and novel perspective. The afternoon, I am part of another defense committee for the habilitation of Fadoua Balabdaoui, my colleague in Paris-Dauphine. Fadoua is working in non-parametric statistics, under shape constraints, but has a wide range of interests and publications that fully justify an habilitation degree at this stage of her career. (Habilitation is a degree required in France and Germany to become a Full Professor and to autonomously advise PhD students.)
