Read in both Le Monde (Sept) and Nature (26 Sept) concerning articles about the poor financial state of many higher education institutions in the UK, if not the most prestigious ones… (Warwick being so far preserved from that issue, if not the neighbouring University of Coventry.) Up to 80% of French universities—prestigious and less prestigious alike—are also facing repeated deficits, for several years in a row, if mostly for different reasons besides the after-effects of COVID and the economical crisis amplified by the Russian invasion.
The main difference (between France and Britain—minus bonnie Scotland!) is in the UK government not providing direct support to higher education institutions, treating them as businesses that must make a profit on average to survive. To add insult to injury, the short-sighted Sunak Government has strongly reduced the number of foreign students (that they considered immigrants) when those are a key component of the universities business models since they pay much higher fees compared with the capped fees for UK nationals. Surprisingly (to me), the new Labour government does not intend to change either policy. The more surprising (to me) when the first institutions to fail will be (are?) those concentrating students from less privileged backgrounds. Quite paradoxical for a party with social ideals and socialist roots. Besides, repeated reductions of the attractivity of Britain for higher education may reach a no-return point in terms of its (World first-tier) scientific productivity.
France is financing its universities for 60% of their budget, while tuition fees in national universities (that do no include Paris Dauphine) are nominal for EU students (and special cases like PhD programs). Progressive cuts have led to tighter budgets, more deficit, with limited alternative sources of financing, since raising students fees is completely taboo. Added to academic positions being little attractive (for offering comparatively low salaries), this abides poorly for France keeping being part of top higher education countries. (It already figures very rarely in the country comparisons produced by Nature on an almost weekly basis.) The new higher education minister is known for pushing towards more “autonomy” (read less financing) of universities and the current State deficit may prove the tipping point for cutting further State support for public higher education institutions at large and fundamental research.
Archive for PhD course
banKrUpt, franc’route
Posted in Kids, pictures, Travel, University life with tags bankruptcy, Brexit, college ranking, Coventry, COVID-19, foreign students, French government, French politics, inflation, Labour, Le Monde, Nature, PhD course, Russian invasion, Scotland, Shanghai ranking, student fees, tuition fees, UK Government, UK politics, UK universities, Université Paris Dauphine, University of Coventry, University of Warwick on November 14, 2024 by xi'anSummer school on Bayesian statistics and computation
Posted in Books, Kids, Statistics, Travel, University life with tags Bayesian computation, Bayesian statistics, Ho Chi Minh City, PhD course, Saigon, summer school, UEH, Vietnam, workshop on July 16, 2023 by xi'an
NCE, VAEs, GANs & even ABC…
Posted in Statistics with tags ABC, Bayesian GANs, CDT, deep learning, energy based model, generative adversarial networks, noise contrasting estimation, normalising constant, normalising flow, partition function, PhD course, Teams, University of Warwick, variational autoencoders on May 14, 2021 by xi'anAs I was preparing my (new) lectures for a PhD short course “at” Warwick (meaning on Teams!), I read a few surveys and other papers on all these acronyms. It included the massive Guttmann and Hyvärinen 2012 NCE JMLR paper, Goodfellow’s NIPS 2016 tutorial on GANs, and Kingma and Welling 2019 introduction to VAEs. Which I found a wee bit on the light side, maybe missing the fundamentals of the notion… As well as the pretty helpful 2019 survey on normalising flows by Papamakarios et al., although missing on the (statistical) density estimation side. And also a nice (2017) survey of GANs by Shakir Mohamed and Balaji Lakshminarayanan with a somewhat statistical spirit, even though convergence issues are not again not covered. But misspecification is there. And the many connections between ABC and GANs, if definitely missing on the uncertainty aspects. While Deep Learning by Goodfellow, Bengio and Courville adresses both the normalising constant (or partition function) and GANs, it was somehow not deep enough (!) to use for the course, offering only a few pages on NCE, VAEs and GANs. (And also missing on the statistical references addressing the issue, incl. [or excl.] Geyer, 1994.) Overall, the infinite variations offered on GANs leave me uncertain about their statistical relevance, as it is unclear how good the regularisation therein is for handling overfitting and consistent estimation. (And if I spot another decomposition of the Kullback-Leibler divergence, I may start crying…)
Jeffreys priors for hypothesis testing [Bayesian reads #2]
Posted in Books, Statistics, University life with tags Arnold Zellner, Bayes factor, Bayesian tests of hypotheses, CDT, class, classics, Gaussian mixture, improper priors, Jeffreys prior, JRSSB, Kullback-Leibler divergence, Oxford, PhD course, Saint Giles cemetery, Susie Bayarri, Theory of Probability, University of Oxford on February 9, 2019 by xi'an
A second (re)visit to a reference paper I gave to my OxWaSP students for the last round of this CDT joint program. Indeed, this may be my first complete read of Susie Bayarri and Gonzalo Garcia-Donato 2008 Series B paper, inspired by Jeffreys’, Zellner’s and Siow’s proposals in the Normal case. (Disclaimer: I was not the JRSS B editor for this paper.) Which I saw as a talk at the O’Bayes 2009 meeting in Phillie.
The paper aims at constructing formal rules for objective proper priors in testing embedded hypotheses, in the spirit of Jeffreys’ Theory of Probability “hidden gem” (Chapter 3). The proposal is based on symmetrised versions of the Kullback-Leibler divergence κ between null and alternative used in a transform like an inverse power of 1+κ. With a power large enough to make the prior proper. Eventually multiplied by a reference measure (i.e., the arbitrary choice of a dominating measure.) Can be generalised to any intrinsic loss (not to be confused with an intrinsic prior à la Berger and Pericchi!). Approximately Cauchy or Student’s t by a Taylor expansion. To be compared with Jeffreys’ original prior equal to the derivative of the atan transform of the root divergence (!). A delicate calibration by an effective sample size, lacking a general definition.
At the start the authors rightly insist on having the nuisance parameter v to differ for each model but… as we all often do they relapse back to having the “same ν” in both models for integrability reasons. Nuisance parameters make the definition of the divergence prior somewhat harder. Or somewhat arbitrary. Indeed, as in reference prior settings, the authors work first conditional on the nuisance then use a prior on ν that may be improper by the “same” argument. (Although conditioning is not the proper term if the marginal prior on ν is improper.)
The paper also contains an interesting case of the translated Exponential, where the prior is L¹ Student’s t with 2 degrees of freedom. And another one of mixture models albeit in the simple case of a location parameter on one component only.
