Archive for Fall

trail de la Sainte Baume [4:02:17, 26.2km, 1700m⁺, 187/517, 172M/395M, 1M6/4M6]

Posted in Kids, Mountains, pictures, Running, Travel with tags , , , , , , , , , , , , , , , , , , on March 16, 2026 by xi'an

A particularly tough first exposure to trail running! I had accumulated quite a decent volume in preparation for this race, with more than 400km in four weeks, but it proved of little impact as I hardly ran, given the nature of the terrain, where passing slower runners was near impossible and where steep slopes forced most participants to walk up (fast) and made me hyper-cautious on the way down. And periodically letting faster runners pass me. Especially after my first fall at kilometre 8 when I bumped into a rock and fell flat (with many scratches but nothing broken). As a result, I went from 141th to 161th to 187th at the successive check-points. The massive differential let me with sore tights the whole week, although not preventing me from jogging every morning. And still the weather was near perfect, cold with no wind, sunny most of the way and no rain till the finish line. The good side is that I did not suffer at all from the hindfoot tendinitis I had developed two weeks earlier. And which soon returned on flat, fast terrains like Ponte de la Liberta!

Particular & paternal congrats to my daughter Rachel, who finished less then 15mn after me and managed her race most strategically, gaining positions at each check-up and ending up 23rd woman! Great race, Ra’!!! And great views during the race that I mostly missed, tunnel visioning the track to avoid yet another fall…

Natural statistical science [#2]

Posted in Statistics with tags , , , , , , , , , , , , , , , on November 23, 2023 by xi'an

A rare occurrence of a Bayesian statistics paper in Nature with this “State estimation of a physical system with unknown governing equations” by Course and Nair. A variational Bayes modelling of a state system observed with noise, but without a physical model on the state (SDE) evolution itself. Which means a prior is set on a non-parametric or neural representation of the drift and a linear approximation is used for the variational approximation, leading to a Gaussian process as the approximate distribution. While this applies to highly complex models, like orbiting black holes, it is somewhat a surprise to meet this application of variational inference in a prestigious general science journal like Nature. (The picture above was taken on the train from Marseille at the end of the Bayes Fall school.)

“The approach is based on a technique called Bayesian inference, which is used widely, but which can be computationally challenging for complex systems.” B. Keith

Familial inference

Posted in Statistics, University life with tags , , , , , , , , , on October 3, 2023 by xi'an

An ISBA-BNP webinar on Wednesday, 4 October, at 17:00 UTC by my friend Steve McEachern:

Familial inference: Tests for hypotheses on a family of centers

Many scientific disciplines face a replicability crisis. While these crises have many drivers, we focus on one. Statistical hypotheses are translations of scientific hypotheses into statements about one or more distributions. The most basic tests focus on the centers of the distributions. Such tests implicitly assume a specific center, e.g., the mean or the median. Yet, scientific hypotheses do not always specify a particular center. This ambiguity leaves a gap between scientific theory and statistical practice that can lead to rejection of a true null. The gap is compounded when we consider deficiencies in the formal statistical model. Rather than testing a single center, we propose testing a family of plausible centers, such as those induced by the Huber loss function (the Huber family). Each center in the family generates a point null hypothesis and the resulting family of hypotheses constitutes a familial null hypothesis. A Bayesian nonparametric procedure is devised to test the familial null. Implementation for the Huber family is facilitated by a novel pathwise optimization routine. Along the way, we visit the question of what it means to be the center of a distribution. The favorable properties of the new test are demonstrated theoretically and in case studies.
This is joint work with Ryan Thompson (University of New South Wales), Catherine Forbes (Monash University), and Mario Peruggia (The Ohio State University).

sticker for birds

Posted in Statistics with tags , , , , on March 3, 2023 by xi'an

A moderately interesting article in the NYT on why birds crash into (my) windows and the limited efficiency of bird stickers, esp. those glued from the inside. Just like insects, who have difficulties understanding the concept of glass (!), birds fly into glass windows because they do not realise they are there. What I find surprising in the case of my own house is that they (fairly rarely) crash into windows of rooms with no perspective, i.e. with a wall behind the window. Birds never enter when these windows are open, for instance. One possible explanation is that under some lighting the window is reflecting the outside for the flying bird rather than the inside of the house. Supporting this explanation is the observation that the crashes I noticed occur around the same time in late Fall, when sun is quite low. (And no connection other than coincidence with yesterday’s post on Thunderbird!)

foliage to the max

Posted in Books, Kids, pictures with tags , , , , , on December 17, 2022 by xi'an

An easy riddle from The Riddler that did not even require coding! Given that a tree changes colours at a random time A distributed according to a Uniform distribution (over (0,1)) and that it sheds its leave at a random time B distributed according to a Uniform distribution (over (A,1)), what is the time when a maximal number of trees show their new colour?

Which means optimising in t the probability that A<t<B. Which is equal to -(1-t)log(1-t) and maximal for t=1-e⁻¹, resulting in a (maximal) fraction of e⁻¹ of the trees holding to their new colour at that time.