Archive for the Mountains Category

the buzz about nuzz

Posted in Books, Mountains, pictures, Statistics with tags , , , , , , , , , , , , , on April 6, 2020 by xi'an

“…expensive in these terms, as for each root, Λ(x(s),v) (at the cost of one epoch) has to be evaluated for each root finding iteration, for each node of the numerical integral

When using the ZigZag sampler, the main (?) difficulty is in producing velocity switch as the switches are produced as interarrival times of an inhomogeneous Poisson process. When the rate of this process cannot be integrated out in an analytical manner, the only generic approach I know is in using Poisson thinning, obtained by finding an integrable upper bound on this rate, generating from this new process and subsampling. Finding the bound is however far from straightforward and may anyway result in an inefficient sampler. This new paper by Simon Cotter, Thomas House and Filippo Pagani makes several proposals to simplify this simulation, Nuzz standing for numerical ZigZag. Even better (!), their approach is based on what they call the Sellke construction, with Tom Sellke being a probabilist and statistician at Purdue University (trivia: whom I met when spending a postdoctoral year there in 1987-1988) who also wrote a fundamental paper on the opposition between Bayes factors and p-values with Jim Berger.

“We chose as a measure of algorithm performance the largest Kolmogorov-Smirnov (KS) distance between the MCMC sample and true distribution amongst all the marginal distributions.”

The practical trick is rather straightforward in that it sums up as the exponentiation of the inverse cdf method, completed with a numerical resolution of the inversion. Based on the QAGS (Quadrature Adaptive Gauss-Kronrod Singularities) integration routine. In order to save time Kingman’s superposition trick only requires one inversion rather than d, the dimension of the variable of interest. This nuzzled version of ZIgZag can furthermore be interpreted as a PDMP per se. Except that it retains a numerical error, whose impact on convergence is analysed in the paper. In terms of Wasserstein distance between the invariant measures. The paper concludes with a numerical comparison between Nuzz and random walk Metropolis-Hastings, HMC, and manifold MALA, using the number of evaluations of the likelihood as a measure of time requirement. Tuning for Nuzz is described, but not for the competition. Rather dramatically the Nuzz algorithm performs worse than this competition when counting one epoch for each likelihood computation and better when counting one epoch for each integral inversion. Which amounts to perfect inversion, unsurprisingly. As a final remark, all models are more or less Normal, with very smooth level sets, maybe not an ideal range

 

on an absurd climbing competition

Posted in Kids, Mountains, pictures, Running, Travel with tags , , , , , , , , , , on April 1, 2020 by xi'an

The New York Times has a very interesting piece on why Adam Ondra, arguably the best sport climber in the World, who climbed the very first 9c route in 2018, with a supernatural move involving hanging head down, actually has little hope of winning the Olympics. Assuming there will be Olympics this year. It is essentially because there is only one single medal for the sport, merging the radically different skills of bouldering, lead climbing and the absurd addition of speed climbing, which involves a single route, always the same, not particularly hard (6b) but to be climbed as fast as possible. To be a top contender on two categories is already pretty rare, with Ondra an exception. To master all three… Only cumulated athletic categories like heptathlon or pentathlon compare, but they come on top of existing competitions for every single of the seven or five events they are made of. Ondra came second or first in bouldering and lead, but closer to last for speed climbing. At least he made it through the qualifications.

IMS-Bernoulli congress delayed [WC2020]

Posted in Mountains, pictures, Running, Statistics, Travel, University life with tags , , , , , , , , , on March 30, 2020 by xi'an

Just received the sad news that the 10th World Congress in Probability and Statistics (WC2020), jointly organized by the Bernoulli Society and IMS, in Seoul, 17-21 August 2020, must be delayed till next year. I expect the same to happen for JSM 2020 in Philly.

ABC in Svalbard [news #1]

Posted in Mountains, pictures, Running, Statistics, Travel, University life with tags , , , , , , , , , , , , , , , , , , , , on March 23, 2020 by xi'an

We [Julien and myself] are quite pleased to announce that

  • the scientific committee for the workshop has been gathered
  • the webpage for the workshop is now on-line (with a wonderful walrus picture whose author we alas cannot identify)
  • the workshop is now endorsed by both IMS and ISBA, which will handle registration (to open soon)
  • the reservation of hotel rooms will be handled by Hurtigruten Svalbard through the above webpage (this is important as we already paid deposit for a certain number of rooms)
  • we are definitely seeking both sponsors and organisers of mirror workshops in more populated locations

As an item of trivia, let me recall that Svalbard stands for the archipelago, while Spitsbergen is the name of the main island, where Longyearbyen is located. (In Icelandic, Svalbarði means cold rim or cold coast.)

the year(s) with no conferences

Posted in Books, Mountains, pictures, Travel, University life with tags , , , , , , on March 21, 2020 by xi'an

This week, Nature has an article on “A year without conferences? How the coronavirus pandemic could change research”, where the journalist predicts a potential halt to scientific conferences. Taking as example the cancelled American Physical Society (APS) March Meeting, to quote

“many of them rapidly set up platforms to hold virtual sessions for the meeting, inviting their speakers to present by webcam or to upload their presentations to online repositories. Researchers who hadn’t been in a position to fly to Denver found themselves able to participate from afar in what became the Virtual APS March Meeting.”

On this same day I should have been traveling from Brussels to Grenoble for the ABC meeting there. Instead, I had a four day virtual panel meeting from home and there is no virtual version of the ABC in Gre[e]noble workshop. As no one seemed particularly eager to animate a few local talks with no guarantee of spectators. As things deteriorated to home confinement,  it was actually better not to spend more efforts on the project. Since this confinement is bound to last much longer, it would however become more obvious that the community and the academic societies need plan virtual conference and invent different channels to gather members and disseminate innovation.

an oldie but a goldie [jatp]

Posted in Mountains, pictures, Travel, University life with tags , , , , , , , , on March 19, 2020 by xi'an

!ABC%in%Grenoble

Posted in Mountains, pictures, R, Statistics, Travel, University life with tags , , , , , , , , , on March 9, 2020 by xi'an

Most regrettably, the incoming ABC in Grenoble, planned on 19-20 March 2020 is now postponed till a yet unspecified date, like next Summer or next Fall. Due to travel disruptions all over the World and to gathering restrictions in France, courtesy of the Covid-19 virus. Not particularly surprising given the avalanche of cancellations, but unfortunate. The mirror meetings that had emerged in the past weeks will most likely be cancelled as well, unless a joint program can be constructed based on the local attendees.

Here is the full message from the organisers to the registered participants:

It is with regret that we have to cancel the ABC in Grenoble workshop. We (local organizers) will be waiting for the situation to settle down in order to possibly reschedule for a future date.  If so, we would open up the call again both for contributed talks and posters.
As you may know, we were already planning to provide electronic options to hold virtual mirrors for the workshop. However, we consider that relying only on streamed talks from various places without a main venue here in Grenoble would make it hard to follow.
This may take some time, but we will let you know as soon as we have further information. Please consult the workshop website for up-to-date information