Archive for poster session

[o-l] ISBA 2021 [百花齊放,百家爭鳴]

Posted in pictures, Statistics, Travel, University life with tags , , , , , , , , , , , , , , on January 20, 2021 by xi'an

As usual, when looking for our ISBA World conference, ISBA [now] 2021, many alternative realities emerge, ISBA being such a popular acronym! The actual dates for the (true and only) ISBA 2021 conference are 28 June – 02 July and the location is no longer Kunming, China, since there is a pandemic going on!!! The conference is now on-line, which forces a complete rethinking of its organisation, away from a sheer replica on Zoom of the traditional conference. And with no registration fees!! I have agreed to join other ISBA members to contribute (with limited abilities) to this switch and obviously welcome comments towards it! Having greatly learned from the earlier experience with the One World Bernoulli+IMS conference last summer, thanks to the dedication and imagination of Leif Döring!, here are some line of thought:

  • Keep in mind days are 24 hours long and attention span much smaller, towards keeping the offer manageable for a fully engaged participant
  • Account and take advantage of the multiple time zones available to virtual participants to stretch the schedule to cover as many participants as possible, with a potential multiplication of (plenary) talks and posters
  • Avoid filling the on-line schedule with live talks but have them pre-registered, possibly with several levels of length and depth, including a one-slide two-minute version
  • Exploit on-line abilities to focus solely (?) on interactions, which is the main point of conferences, meaning participants joining for a thematic session over a reasonable duration to discuss talks or posters they had attended on their own before, with a catalyst leading the show with prepared questions
  • Preregister poster presentations as well, so that live poster sessions involve only questions and discussions, and group poster by theme (an attempt of mine at the earlier ISBA conferences) so that presenters and visitors can interact at the theme level rather than being stuck in an empty room with one’s poster
  • Create local mirrors when people could physically (safely!) gather to attend the conference, from watching videos and poster together to engage into a local plus virtual discussion during interaction sessions. I certainly plan to hold one such session in Paris [if there and not in Eindhoven on 28 June – 02 July]
  • Find ways to engage participants to fully commit to the conference (see above), for instance by preparing a “I am away at ISBA 2021” card one could post on whatever social or asocial networks one favours (but maybe not as one’s vacation email automated reply, unless the card is a few bytes…)

BayesComp 2020 at a glance

Posted in Statistics, Travel, University life with tags , , , , , , , , , , , , , , , , on December 18, 2019 by xi'an

BayesComp 20 [schedule]

Posted in Books, Kids, pictures, R, Statistics, Travel, University life with tags , , , , , , , , , , , on November 20, 2019 by xi'an

The schedule for the program is now available on the conference webpage of BayesComp 20, for the days of 7-10 Jan 2020. There are twelve invited sessions, including one j-ISBA session, and a further thirteen contributed sessions were selected by the scientific committee. And two tutorials on the first day. Looking forward seeing you in Florida! (Poster submissions still welcomed!)

O’Bayes 19/3.5

Posted in Books, pictures, Travel, University life with tags , , , , on July 3, 2019 by xi'an

 


Among the posters at the second poster session yesterday night, one by Judith ter Schure visually standing out by following the #betterposter design suggested by Mike Morrison a few months ago. Design on which I have ambivalent feelings. On the one hand, reducing the material on a poster is generally a good idea as they tend to be saturated and hard to read, especially in crowded conditions. Having the main idea or theorem immediately visible should indeed be a requirement, from immediately getting the point to starting from the result in explaining the advances in the corresponding work. But if this format becomes the standard, it will become harder to stand out! More fundamentally, this proposal may fall into the same abyss as powerpoint presentations, which is that insisting in making the contents simpler and sparser may reach the no-return point of no content [which was not the case of the above poster, let me hasten to state!]. Mathematical statistics poster may be automatically classified as too complicated for this #betterposter challenge as containing maths formulas! Or too many Greek letters as someone complained after one of my talks. And treating maths formulas as detail makes them even smaller than usual, which sounds like the opposite of the intended effect. (The issue is discussed on the betterposter blog, for a variety of opinions, mostly at odds with mine’s.)

O’Bayes 19/3

Posted in Books, pictures, Statistics, Travel, University life with tags , , , , , , , , , , , , , , , on July 2, 2019 by xi'an

Nancy Reid gave the first talk of the [Canada] day, in an impressive comparison of all approaches in statistics that involve a distribution of sorts on the parameter, connected with the presentation she gave at BFF4 in Harvard two years ago, including safe Bayes options this time. This was related to several (most?) of the talks at the conference, given the level of worry (!) about the choice of a prior distribution. But the main assessment of the methods still seemed to be centred on a frequentist notion of calibration, meaning that epistemic interpretations of probabilities and hence most of Bayesian answers were disqualified from the start.

In connection with Nancy’s focus, Peter Hoff’s talk also concentrated on frequency valid confidence intervals in (linear) hierarchical models. Using prior information or structure to build better and shrinkage-like confidence intervals at a given confidence level. But not in the decision-theoretic way adopted by George Casella, Bill Strawderman and others in the 1980’s. And also making me wonder at the relevance of contemplating a fixed coverage as a natural goal. Above, a side result shown by Peter that I did not know and which may prove useful for Monte Carlo simulation.

Jaeyong Lee worked on a complex model for banded matrices that starts with a regular Wishart prior on the unrestricted space of matrices, computes the posterior and then projects this distribution onto the constrained subspace. (There is a rather consequent literature on this subject, including works by David Dunson in the past decade of which I was unaware.) This is a smart demarginalisation idea but I wonder a wee bit at the notion as the constrained space has measure zero for the larger model. This could explain for the resulting posterior not being a true posterior for the constrained model in the sense that there is no prior over the constrained space that could return such a posterior. Another form of marginalisation paradox. The crux of the paper is however about constructing a functional form of minimaxity. In his discussion of the paper, Guido Consonni provided a representation of the post-processed posterior (P³) that involves the Dickey-Savage ratio, sort of, making me more convinced of the connection.

As a lighter aside, one item of local information I should definitely have broadcasted more loudly and long enough in advance to the conference participants is that the University of Warwick is not located in ye olde town of Warwick, where there is no university, but on the outskirts of the city of Coventry, but not to be confused with the University of Coventry. Located in Coventry.