Archive for SNCF

X Fall²⁶ World Tour

Posted in Running, Statistics, Travel with tags , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , on September 21, 2026 by xi'an

As you may have noticed from previous entries, over the past three weeks I have been almost constantly travelling, with various but mostly large amounts of luck in my travels.

Starting with a Paris-Marseille round trip for our daughter birthday, where both trains ran on time with operational WiFi and Air conditioning (and the inevitable noisy junior passengers commuting back from their vacations).

Followed a few hours later by a Paris-Lugano round trip where again the stars aligned since the flight to Milano Malpensa was on time and landed 20mn before the only direct (flexi)bus to Lugano departed. Again in time. On the way back I had to leave right after my talk (apologies!) but got a direct (if slow) train to Milano and hopped the push via Montenapoleone (!) between two Milano metro lines to reach Milano Linate with one hour to go. On time again.

One day at home to sort various pump issues (from fixing all car injectors to replacing a drainage pump that failed to stop a massive rainstorm from entering our basement, to a flat bike tyre returning from PariSanté Campus) and then a quick weekend drive to Argentan, Normandy, for which the car managed and the traffic was surprisingly smooth for the last weekend of the national August vacation break), and where I enjoyed a long run in the woody hills, one month or so prior to the Argentan half-marathon, 30 years from my first participation. Enough time back home to cook a fresh, Norman, rhubarb preserve and off to to Durham, Northumbria, by another smooth, on-time, flight, with the small hitch of the RER B—just back from being closed for an entire month—unexpectedly finishing one station too early (and a particularly noisy family returning from one amusement park). And the same on the return, via Newcastle with a glimpse from the train of the Tyne and its iconic bridge. Reminding me of my visit there for the RSS meeting in 2013, travelling with Jim Hobert.

One more day at home and at work (teaching my first Monte Carlo class of the semester) and I was back in Charles de Gaulle airport for the flight to Tokyo Haneda. With enough time to have a second breakfast in the lounge and leave my phone on my seat!, fortunately recovered a few frantic minutes later. The flight took one hour less than expected, thanks to favourable winds, with the hitches of having no AC power plug available on the plane and no Wifi on either phone or laptop, a nuisance on a 12h day flight with a talk to polish. Going through customs at Haneda was even smoother than “usual” and I only took one wrong train on my way to Yokohama. (Which happened again the next days until I decided to stick Google map schedules, which are highly accurate and constantly updated). The only hitch of the trip was the attempt at bullet summiting Fuji-san, unsurprisingly as with most of my bullet attempts!, because the weather conditions were horrendous to the point of wardens turning people off at the entry of the Yoshida path and also to the point of inducing landslides on the unique route conducting to the 5th station. I thus got stuck there with 1600 people for a few hours until the route was reopened and the shuttle restarted, obviously with some further delays in entering the (third) bus, with more confusion and jostling than usual in Japan (but also with a large proportion of foreign hikers!). Taxis were not responding either and before the news came of the reopening I was getting worried about missing my flight! While the return saw no seamless connections and no express trains, leading to a total trip time close to 5 hours, incl. a cold soba stop at a station, I still managed to reach my Haneda hotel at check-in time and to have an intense training session on their treadmill to make up for the missing 15km and 1400⁺ climb. (Rescheduling my direct bus ticket on-line proved impossible from my phone without a local help as automated translation does not work with images.) The flight back to Paris went very smoothly by the Eastern route over the pole, with AC power (making my external battery delivered by amazon.jp fortunately superfluous since it is prohibited to use on AF flights) and Wifi running. Making my working on the paper associated with the Akaike lecture way more efficient.

off to [ML@]Aussois

Posted in Books, Kids, Mountains, Running, Statistics, Travel, University life with tags , , , , , , , , , , , , , , , , , , on November 14, 2025 by xi'an

 

BayesComp 2025.4

Posted in pictures, Running, Statistics, Travel, University life with tags , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , on June 21, 2025 by xi'an

The third and final day of the (main) conference started tih Emtiyaz Khan’s plenary talk on adaptive Bayesian intelligence. Or, imho, [adaptive [Bayesian]] intelligence, with the brackets indicating redundancy since intelligence need include adaptivity and [intelligent] adaptivity need proceed in a Bayesian way! Focussing first on the Bayesian learning rule via variational Bayes (with a stress on Kingma’s 1994 Adam optimisation algorithm, the “most cited paper” [in machine learning]) where learning boils down to gradient steps (due to the exponential family structure), themselves versions of Taylor (or Laplace) approximations). With an interesting vision of Bayesian updating as accounting for prediction mismatch. (I missed the connection Roberta in IMDb appearing in one slide!)

 The following session offered no dilemma [sorry, Alex, Axel, Chris, Robert, Sumeet, Victor!] since it included the federated learning session I organised, with Louis Asslet, Conor Hassan, and Jean-Michel Marin as speakers. Louis’ talk was on confidential [homomorphic] accept-reject algorithms to learn from other sources, while preserving (differential?) privacy, part of which came during Les Houches workshops I organised this Spring and the one before. Exploiting the additive features of log-likelihoods and exponential variates and adopting a testing perspective on privacy. Conor motivated his model with the Australian cancer atlas project Kerrie Mengersen and others have been developing over the years. The federated approach relies on variational approximations that return the same answer as an exact resolution, but more efficiently. (From a privacy perspective, I wonder at the impact of variational approximations on protecting the data, which boils down to a choice of (sufficient) statistics for the exponential families behind those approximations.) For more complicated models incorporating spatial dependence prohibits full Bayesian inference, unfortunately. Jean-Michel commented on the richness of methods for simulation-based inference, incl. model choice. His focus was on using sequential neural likelihood estimation and sequential importance sampling to approximate evidence. As in the Read Paper of Del Moral et al. (2006). Mentioning a neural version of the harmonic mean estimator by Spurio Mancini et al.  (2023)! I wondered at the degree of (Rao-Blackwell) recycling involved in the computation, Jean-Michel’s answer being that AMIS is soon coming [in a theatre near you!].

The afternoon sessions did offer any reprieve in the choice of topic! I first went to Approximate Methods for Accelerated Sampling, with Rong Tang evaluating the informativeness of summary statistics through a divergence evaluation. Using autoencoders to replace the intractable posterior, with sliced minimal model discrepancy (MMD) and (pseudo?) score matching loss for divergences (reminding me of indirect inference and synthetic likelihood). Yun Yang discussed a variational proposal to estimate the number of components in a mixture model. Surprising given the multimodal structure of mixture posteriors. And the overall irregularity of (evil!) mixture models. But I could not figure out from the talk the form of the approximation.

On the food scene, tasted a nice and spicy Peranakan rice vermicelli dish called Mee Siam yesterday in a campus restaurant, which sustained me fore the rest of the day, including the ABC s/webinar. And another spicy hot pot today at NUS, to catch up on veggies, while missing the chili crab local specialty on that trip.

BayesComp 2025.3

Posted in pictures, Running, Statistics, Travel, University life with tags , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , on June 20, 2025 by xi'an

The second day of the conference started with a cooler and less humid weather (although this did not last!), although my brain felt a wee bit foggy from a lack of sleep (and I almost crashed while running on the hotel treadmill, at 14.5km/h!), and the plenary talk of my friend of many years Sylvia Früwirth-Schnatter on horseshoe priors and time-varying time series (à la West). With a nice closed-form representation involving hypergeometric functions of the second kind (my favourite!), with the addition of a triple-Gamma prior. Sylvia stressed on the enormous impact of the prior choice on change-point detection, which was already the point in the original horseshoe paper (as opposed to George’s Lasso prior). Without incorporating any specific modelling on potential change-point, fair enough given that the parameter is moving with time, unhindered. Her MCMC choices involved discrete parameters with Negative Binomial and Poisson parameters, allowing for partially integrated or collapsed solutions. Possibly further improved by Swendsen-Wang steps.

I then attended the (advanced) Langevin session after agonising upon my choice for a wealth of options! Sam Power presented a talk linking simulation with optimisation targets, over measure spaces. With Wasserstein gradient flow algorithms that resemble Langevin algorithms once discretised by a particle system. (A natural resolution producing a somewhat unnatural form of measure estimator since made of Dirac masses, from which very little can be learned.) Then [my Warwick colleague & coauthor] Any Wang on underdamped Langevin diffusions. when Poincaré‘s inequality fails, but convergence (in total variation) still occurs. Followed by Peter Whalley on splitting methods (where random hypergeometric subsampling dominates Robbins-Monro) and stochastic gradient algorithms, in a connected (to the previous talks) way since involving underdamped aspects. (With a personal discovery of Polyak’s heavy ball method.)

The afternoon session saw me facing a terrible dilemma with three close friends talking at the same time! Eventually opting for PDMPs, over simulation-based inference and recalibration for approximate Bayesian methods. Kengo Kamatani gave a general introduction to PDMPs, before explaining the automated implementation he considered with Charly Andral (during Charly’s visit to ISM, Tokyo, two summers ago). Towards accelerating the generation of the jump time. Then Luke Hardcastle applied PDMPs for survival prediction, using spike & slab priors and sticky PDMPs. And Jere Koskela (formerly Warwick) extended zig-zag sampling to discrete settings (incl. Kingman’s coalescent.)

The (rather long) day was not over yet since we had planned an extra on-site OWABI seminar & webinar with two participants in the conference, Filippo Pagani (Warwick and OCEAN postdoc) using fusion for federated learning, with a trapezoidal approximation, and Maurizio Filippone on GANs as hidden perfect ABC model selection, a GAN providing an automatic density estimator… With astounding Gemini-generated cartoons! Videos are soon to be available. A big congrats to the speakers who managed to convey their ideas and results despite the late hour! (On the extra-academic side, I was invited last night to a genuine Szechuan dinner in Chinatown, with a large array of spicy dishes if not that spicy!, and a rare opportunity to taste abalone. And bullfrogs. Quite a treat! And a good reason to skip dinner altogether!)

BayesComp 2025.1

Posted in Running, Statistics, Travel, University life with tags , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , on June 18, 2025 by xi'an

Minus one day at BayesComp 2025! As I am attending the model misspecification satellite workshop (ten minutes late, due to repeated path finding protocol!), with an extended presentation by Jeremias Knoblauch on post-Bayesian inference, incl. powered likelihood and Gibbs posteriors. A very smooth and pedagogical presentation, esp. in the hybrid mode. A perspective I associate with the difficulties of making sense of the post-posterior, not truly a posterior, of calibrating the penalty (eg λ), picking the loss (α, β, γ divergences?) , and the drift towards learning goals since the new measure is the post-posterior predictive. Sort of paradoxical return to a Gaussian post-posterior on the parameter that does not seem to stay robust. Horrendous computational issues, when the loss itself is an integral. Use of the zig-zag sampler with an estimated unbiased gradient of the loss, much faster than pseudo-marginal, which (naïvely?) makes sense both because PDMPs directly use scores and because of the power of stochastic gradient methods. Worse perspectives for optimisation-centric posterior that are essentially vamped versions of GANs. For instance, what is the meaning of the coverage probabilities?

The second talk by Jonathan Huggins was on DC (not bagged) posteriors as martingale posteriors with m<∞ (approximating marginal distributions with random kernel MCMC—which persists in simulating the marginalised or integrated variable u from its prior, rather than adapting to the current value of the parameter θ— or subsampling MCMC akin to stochastic gradient Langevin) with connection with cut posteriors,

Then I skipped to the second workshop on Bayesian methods for distributional and semiparametric regression, to listen to my friend David Rossell’s talk on local variable selection. Which suffers more than in standard models under misspecification. Another talk involving cut posteriors, the cuts being on the spline bases…

The day and the workshop concluded with great talks by (my friends) Pierre Alquier and David Frazier. David centred his misspecification talk on cut posteriors. Managing to bring in shrinkage estimators (and mention Bill Strawderman!).

A wee stressful trip, since the races in Caen cancelled all buses and delayed the taxi enough to miss the train to Paris by 30s, catching the next available one leaving me less than one hour between the arrival of the train (delayed by construction work on the rail line) and boarding the flight at Charles de Gaulle airport, but fortunately the RER trains in Paris were running okay, there were no queues in the airport, and I thus made it in time with a bit of post-marathon jogging! (Only to be delayed at departure by one hour for stormy conditions over Germany and Austria). All this exercise proved helpful to sleep soundly and lengthily in the plane!