Following my Akaike Memorial Lecture, I was treated to a very traditional kaiseki dinner near Shōmyōji Temple in Kanazawa, Yokohama. The place felt as if there was only room for a single party and the taxi was uncertain he could drive all the way there as the street was getting narrower and narrower!
The dinner consisted of nine dishes, most of which I had not tasted before. Including the one below with abalone and the Japanese equivalent of bottarga. Fabulous experience!
Read Patrick Dewdney’s La Maison des Veilleurs, the fourth volume in this fantastic fantasy series. Only to realise by the end that this was not the end! The setting is the same as in the previous volumes, which makes this one less exciting and prone to extended lazy episodes. The main character, Syffe, keeps puzzling about his destiny and culpability in past events, and it took me a while to get re-hooked, but this remains a remarkable creation. Hopefully soon to be completed! During my Singaporean trip, I also read Sayaka Murata’s Earthlings which is a short fantasy in the Murakami sense, a disrupting mixture of realism (and nostalgia of childhood days) and of unreal behaviours of the characters. The story revolves around mental and sexual aggressions of the main character as a pre-teen, from a paedophile teacher as well as an abusive mother and a transparent father. The story gradually turns away from any pretence to realism and the end is a wee bit of a disappointment. Still worth reading imho. As well as Graham Greene’s The Man Within, which I had not read before (during an episode of intense exploration of his work, following the discovery of The End of the Affair and Brighton Rock). This is one of the earliest Greene’s, and it does contain many hints and premises of the great books to come, with the themes of pursuit, guilt, salvation already there, but as such, the story is rather awkward and contrived, with unnatural dialogues, inner thoughts revolving in circles, and a complete lack of realism in the scenario. The style itself is definitely Greene’s, with beautiful descriptions of the landscape. (I wonder why the cover features the famous Preikestolen cliff on the Lysefjord rather than the more sedate Shoreham downs, where smuggling French alcohols makes sense!)
Made no cooking for an entire week but enjoyed the almost infinite variety of Singaporean cuisines, with some nice discoveries. If missing the chili crab session on two consecutive nights! (And, coïncidence!, sat next to the same passenger on both flights between Paris and Singapore. If not the same crew as it happened once between Amsterdam and Calgary.) Had a weekend to recover and make most of my “classics”, incl. the weekly whey buckwheat galettes. And missing once again to keep an aquafaba mousse mouss-y enough!
Watched one episode of When Life Gives You Tangerines, which features Jeju women (haenyeo) diving for abalone harvesting (which I tasted in Singapore, if not likely from Jeju). And (at last) Oppenheimer on my flight back, as a way to fight sleep towards resetting my internal clock. Which I mostly enjoyed, despite some awkwardness in the scenario and an old-fashioned way of flipping between different eras, with black and white episodes in case one does not notice. And a very poor treatment of female characters. The ghastly moments when the team celebrates the destruction of Nagasaki and Hiroshima or when officials refuse to drop from the nuclear arm race are quite strong, especially these days. Among the MCMC originators, only Edward Teller is involved in the story, from his obsession with the H bomb to his testimony against Oppenheimer at the high of the McCarthy Red Scare… I also sped through Companion, a lame movie about support robots turning into killers that offers little in either thriller or comic perspectives. Possibly written by ChatGPT?!
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!)
 | Ewan on Tidy up, Claude! |
 | xi'an on Tidy up, Claude! |
| Ewan on Tidy up, Claude! |
| Ewan on Já við viðræðum. Já við Evrópu… |
| xi'an on Магучіх синій та золотий [B… |