Entries of these two January issues on
– yet another 100th anniversary!, namely the founding paper of Heisenberg’s quantum mechanics paper in Zeitschrift für Physik, written on the Instagrammable island of Heligoland. With celebrations at the UNESCO in Paris, Anaheim (CA), Kumasi (Ghana), and Salvador de Bahia (Brazil)!;
– some bird species decorating their nest with shed snakeskins, to frighten predators;
– climate predictions for Trump 2.0, albeit the only certainty being it will get worse and worse (and only the beginning of a flow of articles on the Trumpian attacks on science and scientists);
– the curious plight of the open access journal elife loosing its impact factor after getting too open for Clarivate and then seeing submission from China falter;
– the first European cities from 6000 years ago being found in Ukraine (and Romania) within the Cucuteni-Trypillia culture, with apparently an equalitarian structure with no temple or elite, although the lack of written documents (besides beautiful clay figures whose style evolved over the period);
- an AI tool that interprets spreadsheets (which I would find most useful, given my distaste of said spreadsheets!);
and how to help lab workers facing substance disorders. And break
the attached taboos. Which reminded me of a colleague in that situation when I was head of a lab, years and years ago. And of the difficulty of handling the case all by myself…
Also
– a quick report on a Physics Review Letters paper about simulating elections results to spot whether or not margins of victory were properly distributed, having a distribution (for the scaled margins) mostly depending on voter turnout, with a universal shape! The concept behind this analysis is one of universality borrowed from statistical physics. (The fit does not work for the Ethiopian election of 2010 and the Belarus elections during 2004–2019, no wonder!);
– another book review of David Spiegelhalter’s Art of Uncertainty, by Yongyi Min, a statistician from the UN Statistics Division;
creating a DNA base for identifying children kidnapped during a conflict, as for the 20,000 Ukrainian children forcibly deported to Russia over the past three years;
a substantial survey article on neuromorphic computing (submitted in 2023!), about new forms of computing based on hardware/software co-design;
with more technologies to watch in 2025, and another long article (and the cover) on the strong impact of small-scale fisheries on sustainable development.
Yesterday a arXival by Antoine Luciano, Charly Andral (both PhD students, now or then, at Paris Dauphine), Robin Ryder (formerly at Paris Dauphine, now at Imperial College London) and myself got posted. It proposes to improve the scalability of ABC methods by exploiting the (full or partial) exchangeability in the data by implementing permutation-based matching between observed and simulated samples. This significantly improves computational efficiency, which is further enhanced by sequential strategies such as over-sampling, which facilitates early-stage acceptance by temporarily increasing the number of simulated compartments, and under-matching, which relaxes the acceptance condition by matching only subsets of the data. The map of France appears in connection with an application of the method to estimating SIR parameters, department by department. (It is also reminding me of the cover of Markov Chain Monte Carlo methods in practice, the 1996 contributed book edited by Wally Gilks, Sylvia Richardson and David Spiegelhalter.)
