BNP13

BNP13 is set in this incredible location on a massive lake (almost as large as Lac Saint Jean!) facing several tantalizing snow-capped volcanoes… My trip from Paris to Puerto Varas was quite smooth if relatively longish (but I slept close to 8 hours on the first leg and busied myself with Biometrika submissions the rest of the way). Leaving from Paris at midnight proved a double advantage as this was one of the last flights leaving, with hardly anyone in the airport. On Sunday, I arrived early enough to take a quick dip in Lake Llanquihue which was fairly cold and choppy!

Overall the conference is quite exhilarating as all talks are of interest and often covering on-going research. This may be one of the most engaging meetings I have attended in the past years! Plus a refreshing variety of topics and seniority in the speakers.

To start with a bang!, Sonia Petrone (Bocconi) gave a very nice plenary lecture in the most auspicious manner, covering her recent works on Bayesian prediction as an alternative way to run Bayesian inference (in connection with the incoming Read Paper by Fong et al.). She covered so much ground that I got lost before long (jetlag did not help!). However, an interesting feature underlying her talk is that, under exchangeability, the sequence of predictives converges to a random probability measure, a de Finetti way to construct the prior that is based on predictives. Avoiding in a sense the model and the prior on the parameters of that process. (The parameter is derived from the infinite exchangeable [or conditionally iid] sequence, but the sequence of predictives need be defined.) The drawback is that this approach involves infinite sequences, with practical truncation to a finite horizon being an approximation whose precision / error may prove elusive to characterise. The predictive approach also allows to recover a limiting Normal distribution (not a Bernstein-von Mises type!) and hence credible intervals on parameters and distributions.

While this is indeed a BNP conference (!), I was surprised to see lot of talks paying attention to clustering and even to mixtures, with again a recurrent imprecision on the meaning of a cluster. (Maybe this was already the case for BNP11 in Paris but I may have been too busy helping with catering to notice!) For instance, Brian Trippe (MIT) gave a quick intro on his (AISTATS 2022) work on parallel MCMC with coupling. As unbiased MCMC strongly improving upon naïve parallel MCMC relative to the computing cost. With an interesting example where coupling is agnostic to the labeling of random partitions in clustering problems, involving optimal transport, manageable in O(K³log(K)) time when K is the number of clusters.

a journal of the plague year [long weekend reviews]

Read Thinblade, ordered by mistake as I confused the author David Wells for another more famous one! An absolute disaster, from the poor quality of the printed-on-order self-published amazon-made copy to the abyssal style of the author (or of his dog). The story has no depth and no originality [a teenager discovers he must save the World against an evil entity released from captivity and gathers a team of un.be.lie.va.ble followers], the characters are uni-dimensional, either unbelievably good or complete evil and a colour comes with them to tell the hero which is which. The style (or lack thereof!) is massively indigest, with numerous repetitions about the feelings and questionings of the central characters, plus an hilarious focus on food, all menus being included in the text!, same endless drones about the incredible beauty of the visited castles a few days of ride from the hero’s farm. The plot is, again, laughably simplistic, making the Shannara books I read a few months ago sounding like an elaborate literary construct, and completely predictable. I cannot imagine myself or anyone else’s dog reading further books in the series…

Watched The Old Guard after an exhausting day, including a (physical) trip to a dreaded DIY store!, after reading a somewhat lukewarm review in The New Yorker… I found out later that the film is based on a comics series with the same title. And it shows from the lack of real plot (need to get quickly to Afghanistan? just drop out from a freight train in the middle of Sudan…) to the predictability of the story (set-up heroes fight bad guys and at the end, guess what, …), to the massive amount of stale gun fights with the addition of archaic weapons (to make sure everyone understands the old guard is really old!). The funniest part is actually taking place in Goussainville, France, in the ghost section of this town located on the path of De Gaulle airport planes (and thus evacuated, but not demolished), and in its Roman church (listed, hence intact!). The lack of moral imperative or of higher being driving such immortal killers, who mostly seem tired of said immortality, and the absence of connection with the locals (as e.g. in the scenes taking place in Morocco) do not make this B movie any better. (And the French character definitely has an English accent!)

Had a chance lunch in a Michelin recommended restaurant, on the road to Chenonceau and a family vacation, as we were looking for an open restaurant. The haddock appetizer was fantastic (and enough!), while the trout was not so great, presumably frozen, even though the vegetables were original (incl. chayotte) and yummy.

Read Konungsbók (The King’s Book) by Arnaldur Indriðason, found on my mother’s bookshelves, which is a stand-alone book more of the “involuntary spy” type found in Eric Ambler‘s stories than the usual social theme detective story favoured by Indriðason. While the two involuntary spies in the story are indeed two archeolinguists blundering their way through implausible situations, against hidding Nazis and East German police, as Ambler’s The Dark Frontier, the appeal of the book is in the quest for the ultimate Icelandic saga that would close the nation’s history, The King’s Book, towards recovering other foundational and historical documents hoarded by Denmark. At some point, Halldór Guðjónsson Laxness gets the Nobel Prize in Literature, which first stuns the characters into stupefied pride and second helps them into making another unlikely escape. What I enjoyed in the novel is the feeling of ultimate importance attached to the sagas and their role in cementing Iceland as a genuine nation (again connecting with Laxness, whose books described the social desagregation produced by the American occupation).

appel du 18 juin

O’Bayes in action

My next-door colleague [at Dauphine] François Simenhaus shared a paradox [to be developed in an incoming test!] with Julien Stoehr and I last week, namely that, when selecting the largest number between a [observed] and b [unobserved], drawing a random boundary on a [meaning that a is chosen iff a is larger than this boundary] increases the probability to pick the largest number above ½2…

When thinking about it in the wretched RER train [train that got immobilised for at least two hours just a few minutes after I went through!, good luck to the passengers travelling to the airport…] to De Gaulle airport, I lost the argument: if a<b, the probability [for this random bound] to be larger than a and hence for selecting b is 1-Φ(a), while, if a>b, the probability [of winning] is Φ(a). Hence the only case when the probability is ½ is when a is the median of this random variable. But, when discussing the issue further with Julien, I exposed an interesting non-informative prior characterisation. Namely, if I assume a,b to be iid U(0,M) and set an improper prior 1/M on M, the conditional probability that b>a given a is ½. Furthermore, the posterior probability to pick the right [largest] number with François’s randomised rule is also ½, no matter what the distribution of the random boundary is. Now, the most surprising feature of this coffee room derivation is that these properties only hold for the prior 1/M. Any other power of M will induce an asymmetry between a and b. (The same properties hold when a,b are iid Exp(M).)  Of course, this is not absolutely unexpected since 1/M is the invariant prior and since the “intuitive” symmetry only holds under this prior. Power to O’Bayes!

When discussing again the matter with François yesterday, I realised I had changed his wording of the puzzle. The original setting is one with two cards hiding the unknown numbers a and b and of a player picking one of the cards. If the player picks a card at random, there is indeed a probability of ½ of picking the largest number. If the decision to switch or not depends on an independent random draw being larger or smaller than the number on the observed card, the probability to get max(a,b) in the end hits 1 when this random draw falls into (a,b) and remains ½ outside (a,b). Randomisation pays.

air static

[On an Air France flight for Birmingham, two young French students apparently studying in Warwick kept blathering the entire time, with an utter lack of concern for their surroundings. Note: Les Marseillais is a particularly idiotic reality show on French TV.]

•  …j’ai arrêté de regarder les Marseillais, c’est même pas conscient, tu vois…
• …grave, c’est sûr, moi aussi j’ai arrêté, j’avais trop d’épisodes à rattraper…