Archive for Italy
UTMB, c’est parti/andiamo!
Posted in Kids, Mountains, pictures, Running, Travel with tags Alps, Chamonix, Courmayer, France, French Alps, Hoka One One, Italian Alps, Italy, Orsières, Swiss Alps, Switzerland, trail running, ultratrail, Ultratrail du Mont Blanc, UTMB, Valais, Vale d'Aosta on August 25, 2025 by xi'an
…no el xe un pèse!!!
Posted in Kids, pictures, Travel with tags Austria, Austro-Hungary, Carlo Goldoni, Der Nord Wal, Friuli-Venezia, Gallo-Italic languages, German, Giacomo Casanova, Indo-European languages, Italy, Latin, maps, Old Norse, Republic of Venice, Romance language, tee-shirt, TNF, Venetan, Veneto, Venezia, whales on May 19, 2025 by xi'an
And here is the Venetan version of the nuova fanèla (tee-shirt), as suggested by my friend, co-author, and former PhD student Roberto Casarin, who hosted a fantastic ISBA last summer (and kindly looked after me while in ospedale!). Meaning this is not a fish! (If possibly a whale?!) For a better design Antoine and I had to give up the accent in pèse, much to my dismay. (End of the TNF series!)
…Der Nord Wal, but…
Posted in Books, Kids, pictures, Travel with tags Austria, Austro-Hungary, Der Nord Wal, French Revolution, German, Italy, maps, Napoléon Bonaparte, Napoleonic wars, narwhal, Old Norse, Republic of Venice, tee-shirt, TNF, Treaty of Campo Formio, Venetian, Venezia, whales, Yosemite on May 18, 2025 by xi'an
This new version in TNF memes came to life during my stay in Venezia last summer, as the fish or whale analogy struck me (after so many visits!). The German wording of the North whale afforded a proximity to the original brand without sharing a single word with said brand (and no connection with either the North Wall in Yosemite or the narwhal species, whose etymology link to the Old Norse for corpse-skinned whale). Antoine Luciano improved my early sketch of a three-part simplified map into the above (cutting out Santa Croce and inflating Giudecca) et voilà !
However, using a German motto for the map of Venezia may sound offensive to Venetian ears, reminding them of the 50 years of Austrian occupation—with both Napoléon Bonaparte and Napoléon III involved in the matter—, I consulted locals for another version and…
OWABI@BioInference2025 [29 May]
Posted in Mountains, Statistics, Travel, University life with tags Bardonecchia, Bayesian neural networks, BioInference 2025, Gaussian mixture, Gaussian processes, Italy, mixed effect models, mRNA, OWABI, Piedmont, pseudo-marginal MCMC, simulation-based inference, stochastic differential equation, stochastic epidemic compartment models, webinar on May 13, 2025 by xi'an
The next OWABI webinar is going to be quite special, consisting of two selected talks livestreamed from BioInference 2025, a conference on mathematical modelling and inference on (broadly speaking) biological system, taking place in Bardonecchia, Piedmont. The talks will take place on 29 May, 11am CEST (10am BST). The talks will be streamed on the OWABI MS Team Channel as usual.
1st OWABI Talk: 10-10.30am UK time
Speaker: Andrew Golightly (Durham University)
Title: Accelerating Bayesian inference for stochastic epidemic models using incidence data
Abstract: This work considers the case of performing Bayesian inference for stochastic epidemic compartment models, using incomplete time course data consisting of incidence counts that are either the number of new infections or removals in time intervals of fixed length. The most natural Markov jump process representation of the model is eschewed for reasons of computational efficiency, and replaced by a stochastic differential equation representation. This is further approximated to give a tractable Gaussian process, that is, the linear noise approximation (LNA). Unless the observation model linking the LNA to data is both linear and Gaussian, the observed data likelihood remains intractable. Unlike previous approaches that use the LNA in this setting, two approaches for marginalising over the latent process are considered: a correlated pseudo-marginal method and analytic marginalisation via a Gaussian approximation of the noise model. These approaches are compared using synthetic data with the best performing method applied to real data consisting of removal incidence of Oak Processionary moth nests in Richmond Park, London.
2nd OWABI Talk: 10.30-11am
Speaker: Henrik Häggström (Chalmers University)
Title: Simulation-based inference for stochastic nonlinear mixed-effects models with applications in systems biology
Abstract: We propose a novel methodology for Bayesian inference in hierarchical mixed-effects models. By building on our work, we construct a simulation-based inference (SBI) framework that is highly scalable, where amortized approximations to the likelihood and the parameters posterior are first obtained, and these are rapidly refined for each individual dataset, to ultimately approximate the parameters posterior across many individuals. Unlike the current state-of-art SBI methods, which use neural networks, our approximations are expressed via Gaussian mixture models, leading to easily trainable, parsimonious yet expressive surrogate models of both the likelihood function and the posterior distribution. The methodology is exemplified via stochastic differential equation mixed-effects models to describe translation kinetics after mRNA transfection, however the methodology is general and can accommodate other types of stochastic and deterministic models. We compare our approximate inference with exact pseudomarginal inference and show that our methodology is fast and competitive.
MCMC for Bayesian nonparametric mixture modeling under differential privacy
Posted in pictures, Statistics, Travel, University life with tags Bayesian conference, Bayesian inference, Benedictines, Cini Fundation, ERC Synergy Grant, flawed data, Fundazione Giorgio Cini, ISBA 2024, Italy, Ocean, privacy, San Giorgio Maggiore, satellite meetings, vaporetto, Venezia, Venice on July 21, 2024 by xi'an
At the ISBA-Cini workshop on Artificial intelligence, data sharing and regulation, a satellite of ISBA 2024 that was also an ERC Ocean workshop, located in the magnificent buildings of a former Benedictine Monastery (next to the Palladian church), Stéfano Favaro gave a presentation of the output of his ERC Consolidator Grant “Nonparametric Bayes and Empirical Bayes for Species Sampling Problem”, in which he discussed in particular the corpus of work they produced on Bayesian inference with flawed data. He mentioned in particular a recent paper with Mario Berah and Vinayak Rao on running MCMC for Bayesian non-parametric estimation on privatized (noisy) data.
“…if the original dataset Y is modelled as a realization of a BNP mixture model, then, after marginalizing out the Yi’s, the Zi’s again follow a nonparametric mixture model….”
“…our approach has the flavour of the pseudo-marginal MCMC approach (Andrieu and Roberts, 2009), with the latent Yi’s introduced back into the MCMC state as auxiliary variables to deal with intractable Metropolis-Hastings probabilities…”
“As mentioned in Ju et al. (2022), the efficiency of our algorithms are linked to the special structure of differential privacy…”
The paper starts from a privatized model where the data is perturbed to ensure (standard) differential privacy and only examines the convergence impact of the additional randomness layer on MCMC performances, depending on two versions… One re-simulating the hidden data given everything else and the other close to Neal’s (2000) for this specific hierarchical model. While more involved, the resulting MCMC samplers are not much different from the original ones since the privatization leads to replacing the initial distribution with a convolution, but the actual data still proceeds from an infinite mixture. In that respect, the paper does not engage in a discussion of privacy, like the impact of the BNP estimation, possible gains due to producing a BNP estimator, &tc. The only clear impact of the privacy assumptions is that the Metropolis-Hastings acceptance probability is lower bounded, but this was also noticed in Ju et al. (2022). It would have been nice to assess the impact of non-parametrics on further protecting the data if any!
