Archive for RATP

a journal of the dynamite year

Posted in Books, Kids, Mountains, pictures, Travel, Wines with tags , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , on November 19, 2025 by xi'an

 Read another Fred Vargas novel, Sans feu ni lieu (The Accordionist), a 1997 detective story that is recycling the characters of Debout les morts! (The Three Evangelists). It took me a while to engage with the story since the dialogues are terribly artificial and dated. The resolution of the crime series is also terrible, with an absurd link to a poem by Gérard de Nerval posted in the Paris métro… Also read Orogénèse, a novel written by Dominique Picard—a French statistician at Paris Diderot, who worked on differential geometry and wavelets, and was an ICMS speaker in 2006—, about a reluctant (male) lawyer taking the case of a murderess who does not want to be defended. This book is inspired by the true story of Fabienne Kabou, who drowned her 15 month daughter on a North Sea beach in 2013. And presumably by the book written by her (female) lawyer. Immensely troubling story that is reflected by the style of the novel and the half-truths in the lawyer monologue, Orogénèse stays away from judging or explaining, by slowly exploring the surrounding of the infanticide. Impressive (in many senses).

Made an attempt at a “no-bun burger” using panned Portobello mushrooms—making a very rare appearance on my market stalls—that did not prove very practical as they could not hold much of the onion, avocado and salad contents. But tastier than the (button) Paris mushroom type. We also had an unusual series of take-outs from the local Thaï restaurant, out of solidarity with the long-time owners as their place was about to close and be demolished by the owner. Also made rhubarb compote with the last stems of the year, along with the very few apples that grew in our garden this year.

Watched part of the Korean drama A girl who sees smells (냄새를 보는 소녀). Which is unsurprisingly as terrible as the title suggests. Also watched House of Dynamite after listening to Kathryn Bigelow on the French National Public Radio (France Inter). While the message against [unstable and irreversible] nuclear terror “balance” and for immediate disarmament is necessary, esp. in the current atmosphere of unilateral military actions and of unhinged Supreme Leaders!, the film is terrible and unrealistic, even though The Guardian disagrees. Maybe because the decision chain appears to be less (rather than more) than the sum of the individual decisions. Maybe because of the accumulation of irrelevant actions in the face of the incoming missile. The (US) Centre for Arms Control and Non-Proliferation (CACNP) has provided a detailed analysis of the right and wrong of the scenario, the most obvious for me being the need to hit back at a potential originator (of the nuclear missile) before the missile strikes. (The US’s Missile Defense Agency (MDA) also criticised the failed interception of the attacking missile as unrealistic but CACNP states that the probability of a hit is around 50% , at a cost of $53 billion for 44 interceptors!) That they fail to explicitly name this originator (as North Korea) may be due to fears of cyber attacks as in the earlier instance of The Interview. (The New York Time also enjoys the pace and tension of the movie, while providing a recollection of the best earlier movies acting on the same cliffhanger.)

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!

winter, in altitude

Posted in Kids, Mountains, pictures, Running, Travel with tags , , , , , , , , , , , , , , , on December 22, 2024 by xi'an