Archive for Vancouver

ISBA 2021 grand finale

Posted in Kids, Mountains, pictures, Running, Statistics, Travel, University life, Wines with tags , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , on July 3, 2021 by xi'an

Last day of ISBA (and ISB@CIRM), or maybe half-day, since there are only five groups of sessions we can attend in Mediterranean time.

My first session was one on priors for mixtures, with 162⁺ attendees at 5:15am! (well, at 11:15 Wien or Marseille time), Gertrud Malsiner-Walli distinguishing between priors on number of components [in the model] vs number of clusters [in the data], with a minor question of mine whether or not a “prior” is appropriate for a data-dependent quantity. And Deborah Dunkel presenting [very early in the US!] anchor models for fighting label switching, which reminded me of the talk she gave at the mixture session of JSM 2018 in Vancouver. (With extensions to consistency and mixtures of regression.) And Clara Grazian debating on objective priors for the number of components in a mixture [in the Sydney evening], using loss functions to build these. Overall it seems there were many talks on mixtures and clustering this year.

After the lunch break, when several ISB@CIRM were about to leave, we ran the Objective Bayes contributed session, which actually included several Stein-like minimaxity talks. Plus one by Théo Moins from the patio of CIRM, with ciccadas in the background. Incredibly chaired by my friend Gonzalo, who had a question at the ready for each and every speaker! And then the Savage Awards II session. Which ceremony is postponed till Montréal next year. And which nominees are uniformly impressive!!! The winner will only be announced in September, via the ISBA Bulletin. Missing the ISBA general assembly for a dinner in Cassis. And being back for the Bayesian optimisation session.

I would have expected more talks at the boundary of BS & ML (as well as COVID and epidemic decision making), the dearth of which should be a cause for concern if researchers at this boundary do not prioritise ISBA meetings over more generic meetings like NeurIPS… (An exception was George Papamakarios’ talk on variational autoencoders in the Savage Awards II session.)

Many many thanks to the group of students at UConn involved in setting most of the Whova site and running the support throughout the conference. It indeed went on very smoothly and provided a worthwhile substitute for the 100% on-site version. Actually, I both hope for the COVID pandemic (or at least the restrictions attached to it) to abate and for the hybrid structure of meetings to stay, along with the multiplication of mirror workshops. Being together is essential to the DNA of conferences, but travelling to a single location is not so desirable, for many reasons. Looking for ISBA 2022, a year from now, either in Montréal, Québec, or in one of the mirror sites!

parallel tempering on optimised paths

Posted in Statistics with tags , , , , , , , , , , , , , , , on May 20, 2021 by xi'an


Saifuddin Syed, Vittorio Romaniello, Trevor Campbell, and Alexandre Bouchard-Côté, whom I met and discussed with on my “last” trip to UBC, on December 2019, just arXived a paper on parallel tempering (PT), making the choice of tempering path an optimisation problem. They address the touchy issue of designing a sequence of tempered targets when the starting distribution π⁰, eg the prior, and the final distribution π¹, eg the posterior, are hugely different, eg almost singular.

“…theoretical analysis of reversible variants of PT has shown that adding too many intermediate chains can actually deteriorate performance (…) [while] on non reversible regime adding more chains is guaranteed to improve performances.”

The above applies to geometric combinations of π⁰ and π¹. Which “suffers from an arbitrarily suboptimal global communication barrier“, according to the authors (although the counterexample is not completely convincing since π⁰ and π¹ share the same variance). They propose a more non-linear form of tempering with constraints on the dependence of the powers on the temperature t∈(0,1).  Defining the global communication barrier as an average over temperatures of the rejection rate, the path characteristics (e.g., the coefficients of a spline function) can then be optimised in terms of this objective. And the temperature schedule is derived from the fact that the non-asymptotic round trip rate is maximized when the rejection rates are all equal. (As a side item, the technique exposed in the earlier tempering paper by Syed et al. was recently exploited for a night high resolution imaging of a black hole from the M87 galaxy.)

Thin Air [book review]

Posted in Books with tags , , , , , , , , , , , , , , on April 3, 2020 by xi'an

When visiting Vancouver last December [at a time when traveling was still possible], I had the opportunity to revisit White Dwarf Books [thirty years after my first visit] and among other books bought a Richard Morgan‘s novel, Thin Air, that I did not know existed and which was recommended by the [friendly] book seller. As superior to Morgan’s foray into dark fantasy (that I did not dislike so much). As I had really enjoyed the Altered Carbon series, I jumped on this new novel, which is a form of sequel to Th1rt3en, and very very similar in its futuristic pastiche of tough detectives à la Marlowe, dry humour included. A form of space noir, as The Guardian puts it. I sort of got quickly lost in the plot and (unusually) could not keep track of some characters, which made reading the book a chore towards the end. Thanks to the COVID-19 quarantine, I still managed to finish it, while home cycling!, the very end being more exciting than the beginning drudgery and the predictable sex scenes bound to occur in every of his novels. The Martian world in the novel is only alluded to, which makes it more appealing, despite the invasive jargon, however it sounds too much like a copy of our 20th century with car chase and gun/knife fights. Enhanced by an embedded AI when one can afford it. Certainly not the best read in the series but enough to tempt me into looking at the first episodes of Altered Carbon on Netflix. [Note: the book is not to be confused with the bestselling Krakauer’s Into Thin Air, which relates the 1996 Everest disaster, soon turned into a rather poor movie. I had not realised till today that the same Krakauer wrote Into the Wild…!]

Nature tidbits [the Bayesian brain]

Posted in Statistics with tags , , , , , , , , , , , , , , on March 8, 2020 by xi'an

In the latest Nature issue, a long cover of Asimov’s contributions to science and rationality. And a five page article on the dopamine reward in the brain seen as a probability distribution, seen as distributional reinforcement learning by researchers from DeepMind, UCL, and Harvard. Going as far as “testing” for this theory with a p-value of 0.008..! Which could be as well a signal of variability between neurons to dopamine rewards (with a p-value of 10⁻¹⁴, whatever that means). Another article about deep learning about protein (3D) structure prediction. And another one about learning neural networks via specially designed devices called memristors. And yet another one on West Africa population genetics based on four individuals from the Stone to Metal age (8000 and 3000 years ago), SNPs, PCA, and admixtures. With no ABC mentioned (I no longer have access to the journal, having missed renewal time for my subscription!). And the literal plague of a locust invasion in Eastern Africa. Making me wonder anew as to why proteins could not be recovered from the swarms of locust to partly compensate for the damages. (Locusts eat their bodyweight in food every day.) And the latest news from NeurIPS about diversity and inclusion. And ethics, as in checking for responsibility and societal consequences of research papers. Reviewing the maths of a submitted paper or the reproducibility of an experiment is already challenging at times, but evaluating the biases in massive proprietary datasets or the long-term societal impact of a classification algorithm may prove beyond the realistic.

Metropolis in 95 characters

Posted in pictures, R, Statistics, Travel with tags , , , , , , , , on January 2, 2020 by xi'an

Here is an R function that produces a Metropolis-Hastings sample for the univariate log-target f when the later is defined outside as another function. And when using a Gaussian random walk with scale one as proposal. (Inspired from a X validated question.)

m<-function(T,y=rnorm(1))ifelse(rep(T>1,T),
  c(y*{f({z<-m(T-1)}[1])-f(y+z[1])<rexp(1)}+z[1],z),y)

The function is definitely not optimal, crashes for values of T larger than 580 (unless one modifies the stack size), and operates the most basic version of a Metropolis-Hastings algorithm. But as a codegolf challenge (on a late plane ride), this was a fun exercise.

%d bloggers like this: