Pierre Nyquist
banner
pierrenyquist.bsky.social
Pierre Nyquist
@pierrenyquist.bsky.social
Assoc. prof. in mathematics at Chalmers/University of Gothenburg. Mainly probability, analysis and computational methods, but I like to visit most areas of mathematics. Elected member Young Academy of Sweden. Mostly here for the science, some for the jokes
You managed to live in Boston and not be converted to liking it even a little bit? Strong!
February 9, 2026 at 1:47 AM
intellectual effort were interesting. With the increased use of LLMs to shape courses etc., I found hearing this comforting. Same thing with how everyone that is interviewed highlight and champion their students. Ok, back to caring for the sick people over here.
January 31, 2026 at 9:14 PM
and refreshing to hear from some of the leading researchers in their fields. Sure, they might have arrived at this feeling later on in their careers, when the pressure of publishing and so on is gone, but still. Especially Stephen Boyd's comments on teaching and how it should be an intense
January 31, 2026 at 9:14 PM
spoken at great length on the benefits and joys of teaching, how it is not something to do "on the side" of research etc. This is counter to all the incentives you will typically see at a local level (I am fortunate that it does not really apply to the places I am affiliated with, only some)
January 31, 2026 at 9:14 PM
In particular, it has been wonderful to listen to the different thoughts on teaching and supervision. Of course there is a selection bias—there are of course other equally accomplished scientists that perhaps are not as suited for/will decline a podcast—but invariably so far they have all
January 31, 2026 at 9:14 PM
Currently thinking about starting some kind of activity for grad students and postdocs where we discuss these meta-matters—how to read and write papers, different ways to pick topics, how to approach seminars etc.—and something like this could be an interesting case study in that setting. (8/8)
January 29, 2026 at 11:00 AM
process actually helps make the science better. Not in terms of the actual results, but in how we communicate it. Of course this could happen outside of the journal system, but I do not yet have a good, feasible idea for how that is not "send it to a friend/colleague that gives you comments". (7/8)
January 29, 2026 at 11:00 AM
cumbersome derivations (Taylor expansions galore!). This is now the second paper in a row for Federica where the referee comments really highlighted some blindspots for us and allowed us to make the papers significantly better. Overall, I think it is good to highlight these cases, where the (6/8)
January 29, 2026 at 11:00 AM
time, but in that we were building in a lot of implicit assumptions and interpretations. In hindsight it is clear that the paper is now (much!) better, it is aimed more directly at the intended audience, and the main results are easy to find without going through all the necessary but (5/8)
January 29, 2026 at 11:00 AM
took care to formulate what the essence of the paper was to us. This meant new title, re-writing much of the introduction to change the perspective, re-organise the paper to make it clear what the main results were, etc. Of course we had done what we thought was a good job with this the first (4/8)
January 29, 2026 at 11:00 AM
the reviewers saw it merely as an application of our past work (see below) and the main point was checking a specific condition. After reading through the paper again, it was clear that although that was not how we intended it, that was exactly the way the paper read! We then sat down and (3/8)
A large deviation principle for the empirical measures of Metropolis–Hastings chains
To sample from a given target distribution, Markov chain Monte Carlo (MCMC) sampling relies on constructing an ergodic Markov chain with the target di…
www.sciencedirect.com
January 29, 2026 at 11:00 AM
go at Annals of Applied Probability. I thought it was a 50/50 proposition for AAP, mainly because the reviewers were likely to either like the approach or not, and that would in my mind decide the outcome. When the reviews came back, I was rather surprised by the interpretation of the paper: (2/8)
January 29, 2026 at 11:00 AM
Oh it's been something trying to explain all of this to the biology-people that were also on the paper.
December 20, 2025 at 9:20 PM
Quoting one of the co-authors from our email thread: "After more than four years, it is finally online!" . Especially happy for Federica and Andrei that really drove this project (also, for not having to go through the proofs again to check them)
Sensitivity approximation by the Peano-Baker series
link.springer.com
December 17, 2025 at 9:56 PM
A sad day for the entire Brown community, and of course just an immense tragedy in general. Also shows you how the "but not here"-illusion is just that, an illusion.
December 14, 2025 at 9:43 AM