Botond Szabó gave a BNP webinar last week on the recent paper he wrote with Bocconni colleagues Francesco Pozza and Daniele Durante, to appear in Series B. Which studies the impact of using skew-symmetric approximations of posterior distributions. Skew-symmetric distributions are easy to simulate, either by accept-reject or by exploiting the cdf x pdf structure and the symmetry in the pdf. The Bernstein-von Mises theorem can be expanded to this case, although I am not certain what this means! The main theoretical result is a gain in the magnitude of the approximation, eg in KL, which I did not expected. With questions about the choice of the cdf (which can be automatised when the original posterior is available or when a closed-form approximation replaces it) and of the symmetry point ξ for complex models (which seems to be the MAP by default.) and of the impact on marginal likelihood approximations (if it makes any sense).
Archive for Università Bocconi
Skew-symmetric approximations of posterior
Posted in Statistics with tags Bayesian nonparametrics, BNP Section, Milano, Series B, skew-Normal distribution, skew-symmetric distribution, Università Bocconi, webinar on February 26, 2026 by xi'anBotond Szabó gave a BNP webinar last week on the recent paper he wrote with Bocconni colleagues
mostly Monte Carlo, November
Posted in pictures, Statistics, Travel, University life with tags coupling, f-divergence, MCMC, Monte Carlo Statistical Methods, Paris Sciences et Lettres, PariSanté campus, PSL, seminar, Università Bocconi, Université Paris Dauphine, University of Warwick, variational inference on November 7, 2025 by xi'an
-
Omiros Papaspiliopoulos (Università Bocconi): Bilinear mixed models and variational inference
-
Adrien Corenflos (University of Warwick): A coupling-based approach to f-divergences diagnostics for Markov chain Monte Carlo [recently discussed on the ‘Og]
ERC Starting Grants 2025
Posted in Kids, Travel, University life with tags #ERCSyG, applied mathematics, data science, ERC, ERC Starting Grant, EU grants, EU Horizon Europe programme, EU-Israel Association Agreement, Europe, European Commission, European Research Council, European Union, France, French politics, Horizon Europe, panel, Paris Sciences et Lettres, PE1, PSL, pure mathematics, Switzerland, UK, Università Bocconi, University of Warwick, Ursula von der Leyen on September 13, 2025 by xi'an
Last week, the ERC announced the winners of the Starting Grant campaign for 2025. Congrats to all recipients!
I am no longer a member of the PE1 (maths & data sciences) panel, hence did not contribute in the least to this selection. I noticed very few statistics and data science in the list (albeit obviously not hinting at any correlation with the previous sentence!), except for a project on computational stats. There is neither no one from my departments, even tough Warwick received one grant, nor from my research circles. The re-emergence of the UK as a strong proponent among the nations submitting projects is clear, while Italian candidates remain highly successful (incl. Bocconi!).
Since the EU States did not manage to unanimously agree on suspending the Article 2 of the EU-Israel Association Agreement last summer, despite Israel’s breach of human rights on the Gaza strip, there are also ten Israeli projects supported by EU on this round. (Unless the latest announcement of further sanctions by President Ursula von der Leyen suspending bilateral support to Israel does cover this financing.)
mostly [14] M[ar]C[h] seminar
Posted in Books, Statistics, University life with tags dimension reduction, GLMs, gradient algorithm, importance sampling, Issy-les-Moulineaux, log-normal distribution, MCMC, Monte Carlo methods, Mostly MC, Mostly MCMC seminar, Ocean, Paris, PariSanté campus, Porte de Versailles, PSL, seminar, Università Bocconi, Université Paris Dauphine on March 8, 2025 by xi'an
All about that [Bayes] seminar [24 Jan]
Posted in Books, pictures, Statistics, Travel, University life with tags #ERCSyG, ABC, All about that Bayes, Approximate Bayesian computation, approximate Bayesian inference, Bayesian inference, embarrassingly parallel MCMC, Gibbs sampling, goodness of fit, hit and run algorithm, Institut de Mathématiques de Jussieu, MCMC algorithms, Ocean, Place Jussieu, SCAI, score function, seminar, Università Bocconi, Université Paris Dauphine on January 13, 2025 by xi'an
The next All about that (Bayes) seminar will take place on Friday 24 Jan at SCAI, on the Jussieu campus, with the following talks. (Appearances to the contrary, I was not in the least involved in the program!)
13h30 – 14h30 Joshua Bon (OCEAN, Université Paris Dauphine) – Bayesian score calibration for approximate models
Scientists continue to develop increasingly complex mechanistic models to reflect their knowledge more realistically. Statistical inference using these models can be challenging since the corresponding likelihood function is often intractable and model simulation may be computationally burdensome. Fortunately, in many of these situations, it is possible to adopt a surrogate model or approximate likelihood function. It may be convenient to conduct Bayesian inference directly with the surrogate, but this can result in bias and poor uncertainty quantification. In this paper (https://arxiv.org/abs/2211.05357) we propose a new method for adjusting approximate posterior samples to reduce bias and produce more accurate uncertainty quantification. We do this by optimizing a transform of the approximate posterior that maximizes a scoring rule. Our approach requires only a (fixed) small number of complex model simulations and is numerically stable. We demonstrate beneficial corrections to several approximate posteriors using our method on several examples of increasing complexity.
14h30 – 15h30 Giacomo Zanella (Bocconi University) – Entropy contraction of the Gibbs sampler under log-concavity
In this talk I will present recent work (https://arxiv.org/abs/2410.00858) on the non-asymptotic analysis of the Gibbs sampler, a classical and popular MCMC algorithm for sampling. In particular, under the assumption that the probability measure π of interest is strongly log-concave, we show that the random scan Gibbs sampler contracts in relative entropy, and provide a sharp characterization of the associated contraction rate. The result implies that, under appropriate conditions, the number of full evaluations of π required for the Gibbs sampler to converge is independent of the dimension. If time permits, I will also discuss connections and applications of the above results to the problem of zero-order parallel sampling, as well as extensions to Hit-and-Run and Metropolis-within-Gibbs.
Based on joint work with Filippo Ascolani and Hugo Lavenant.
16h00 – 17h00 Paul Bastide (Université Paris Cité) – Goodness of Fit for Bayesian Generative Models with Applications in Population Genetics
In population genetics, inference about intractable likelihood models is common, and simulation methods, including Approximate Bayesian Computation (ABC) and Simulation-Based Inference (SBI), are essential. ABC/SBI methods work by simulating instrumental data sets of the models under study and comparing them with the observed data set y⁰. Advanced machine learning tools are used for tasks such as model selection and parameter inference. The present work focuses on model criticism. This type of analysis, called goodness of fit (GoF), is important for model validation. It can also be used for model pruning when the number of candidates to be considered is excessive, especially in the context where data simulation is expensive. We introduce two new GoF tests based on the local outlier factor (LOF), an indicator that was initially defined for outlier and novelty detection. We test whether y⁰ is distributed from the prior predictive distribution (pre-inference GoF) and whether there is a parameter value such that y⁰ is distributed from the likelihood with that value (post-inference GoF). We evaluate the performance of our two GoF tests on simulated datasets from three different model settings of varying complexity, and on a dataset of single nucleotide polymorphism (SNP) markers for the evaluation of complex evolutionary scenarios of modern human populations.
Joint work with Guillaume Le Mailloux, Jean-Michel Marin and Arnaud Estoup.
