Archive for Paris Sorbonne Université
a week of causality and explainable [p]A[r]I[s] (20-24 Oct 2025)
Posted in Statistics, Travel, University life with tags AI, causality, explainability, France, Paris, Paris Sorbonne Université, Place Jussieu, predictive models, winter school on September 8, 2025 by xi'an
MaSeMo: Markov, Semi-Markov Models & Associated Fields [Paris, 1-4 July]
Posted in Statistics, University life with tags French universities, Markov chain, Markov model, MaSeMo, Paris, Paris Sorbonne Université, Place Jussieu, semi-Markov chain, workshop on May 7, 2025 by xi'an
Claire Voisin interviewed in Nature
Posted in Books, University life with tags Alexander Grothendieck, algebraic geometry, base, Claire Voisin, CNRS, Crafoord Prize, Hodge conjecture, Institut de Mathématiques de Jussieu, interview, Jussieu, Kodaira problem, Millennium Prize Problems, Nature, Paris Sorbonne Université, philosophy of sciences, René Descartes, set theory on April 8, 2024 by xi'an
Mathematics gets rarely featured in Nature, as this is not the most obvious outlet for mathematical discoveries, hence it is exceptional to see Claire Voisin interviewed in the 08 Feb issue, following her receiving the 2024 Crafoord Prize in Mathematics and Astronomy “for outstanding contributions to complex and algebraic geometry, including Hodge theory, algebraic cycles, and hyperkähler geometry”. She is actually the very first female mathematician to received this prize, although she states this has no particular significance for her.
“In French schools at the time, there was the fashion of ‘modern mathematics’, which was an attempt to teach abstract mathematics, such as set theory. We had to do completely crazy things, like compute the development of numbers in base 2.”
The above statement puzzled me as, given that I am the same age as Claire Voisin, I do not remember doing `modern mathematics’ in high school but rather in primary school, with exposure to sets theory and functions, incl. injectivity and surjectivity, and indeed to different bases, which I found quite fun (rather than crazy) at the time. She also mentioned that, while in preparatory school, she was more interested in philosophy than mathematics, which was almost the case for me as well, in that I would have switched from the maths+physics program there to a program mixing philosophy and maths, had it existed at the time.
All About that Bayes stroll
Posted in pictures, Statistics, University life with tags All about that Bayes, Campus Pierre et Marie Curie, clustering, hidden Markov models, HMM, Institut de Mathématique d'Orsay, Lawrence Livermore National Laboratory, multiple testing, nonlinear ICA, Paris Sorbonne Université, SCAI, seminar, variational autoencoders, youtube on February 9, 2024 by xi'anFor all Bayesians and sympathisers in the Paris area, an incoming All about that Bayes seminars¹ by Elisabeth Gassiat (Institut de Mathématiques d’Orsay) on 13 February, 16h00, on Campus Pierre & Marie Curie, SCAI:
A stroll through hidden Markov models
Hidden Markov models are latent variables models producing dependent sequences. I will survey recent results providing guarantees for their use in various fields such as clustering, multiple testing, nonlinear ICA or variational autoencoders.
¹Incidentally, I came across an unrelated All about that Bayes YouTube video, a talk given by Kristin Lennox (Lawrence Livermore National Laboratory). And then found out a myriad of talks or courses using that pun.
All About that Bayes restart
Posted in pictures, Statistics, University life with tags All about that Bayes, Campus Pierre et Marie Curie, Centrale Supélec, Marie Curie, Paris Sorbonne Université, PCA, Pierre Curie, principal components, Sceaux, seminar, stamp on September 21, 2023 by xi'an
For all Bayesians and sympathisers in the Paris area, All about that Bayes seminars are restarting this semester with a talk by Kaniav Kamari (Centrale Supélec) on 10 October, 16h00, on Campus Pierre & Marie Curie, SCAI:
Bayesian principal component analysis
The technique of principal component analysis (PCA) has recently been expressed as the maximum likelihood solution for a generative latent variable model. In this talk, I’ll first present probabilistic reformulation that is the basis for a Bayesian treatment of PCA. Then, my focus will be on showing that the effective dimensionality of the latent space (equivalent to the number of retained principal components) can be determined automatically as part of the Bayesian inference procedure.
