Sabitlenmiş Tweet
Apparently this is my claim to fame
depths of wikipedia!@depthsofwiki
obsessed with whoever saw gum on a sidewalk and was like "let me add this to the poisson distribution wikipedia article"
English
Moritz Schauer
2.1K posts
Moritz Schauer
@MoritzSchauer
Statistician, Associate professor, Chalmers University of Technology and University of Gothenburg
Katılım Ocak 2018
995 Takip Edilen1.1K Takipçiler
It is even unclear how long the ban of using arxiv as a preprint server extends after: “Penalty is 1 year ban from arXiv followed by a requirement that subsequent arXiv submissions must first be accepted at a reputable peer-reviewed venue”
English
It has “tough on crimes” vibes where to please the hardliners one implements a very crude policy which bites us later. Is this CS or all of arxiv?
Thomas G. Dietterich@tdietterich
Attention @arxiv authors: Our Code of Conduct states that by signing your name as an author of a paper, each author takes full responsibility for all its contents, irrespective of how the contents were generated. 1/
English
@littmath I am producing some interesting solutions to mathematical problems myself but try to prompt me on questions in algebraic geometry…
English
…when I take the exact same prompts that generated these (IMO impressive) solutions and try them on questions in algebraic geometry that I suspect are of comparable difficulty to Erdős problems, they typically produce nonsense (though they can now be helpful with smaller tasks).
English
Still underrated how uneven frontier models are within math, IMO. I’ve recently been reading through some of the more interesting solutions to Erdős problems and quite enjoying them—here the models are reliably executing nontrivial ideas, combining known techniques, etc. But…
English
Moritz Schauer retweetledi
A new letter in the Journal of Diabetes, Science and Technology (£) showing that 40-45% of infants in neonatal units have "impossible" high insulin to C-peptide ratios. This letter will potentially have important implications for the Lucy Letby case. journals.sagepub.com/doi/10.1177/19…
English
Moritz Schauer retweetledi
I want to briefly talk about the use of AI in this paper, which I think is the first paper I've written where LLMs had a substantial effect on the final product, as you can perhaps tell from the acknowledgments:
Daniel Litt@littmath
New paper just dropped, joint with Thomas Krämer and Marco Maculan. It's about a (somewhat mysterious, to me) connection between cubic threefolds and the exceptional Lie group E_6. 1/n
English
Moritz Schauer retweetledi
With the Boston Marathon today, a good time to re-up one of the greatest figures in sports data: the distribution of marathon finish times (n=9,789,093). The spike at 4:00 is not a coincidence.
English
Moritz Schauer retweetledi
GPT-5.4 Pro solves Erdős Problem #1196!
Very pleased with this result; definitely my favourite thus far! This problem has been thought about for some time which makes this reasonably impressive and meaningful (see Lichtman's comments below).
Formalisation is underway!
English
Here it still works: the fly and the acorn are the human tells
Spencer Greenberg 🔍@SpencrGreenberg
I asked Twitter if they/you could tell which of these still lifes were made by AI and which were painted by a 19th-century painter. Here's what happened... 🧵
English
Yeah most nice/good/healthy things are somewhat correlated - the g-factor of life
Matthew B. Jané@MatthewBJane
I’m coming back once again to defend scatter plots. My global response to the skeptical comments: - it’s a small correlation. People don’t seem to like the idea that small correlations simply exist in nature. They do and it’s fine. Are small correlations useful? Maybe sometimes, sometimes not, it depends on the context and application. - it’s definitely not zero. - It’s not p-hacked. Way too much power to have occurred from spurious significance via multiple testing.
English
If you want to look over my shoulder doing math with LLMs here is me instructing ChatGPT to (re-)prove a core argument from our paper Pelekis and Schauer, 2013, link.springer.com/chapter/10.100…
youtube.com/watch?v=EswNPD…
YouTube
English
Our ”Compositionality in algorithms for smoothing” published in Compositionality: doi.org/10.46298/compo… 🎉
Moritz Schauer@MoritzSchauer
A paper that started with a tweet, now its submitted: Compositionality in algorithms for smoothing / Moritz Schauer, Frank van der Meulen, Andi Q. Wang arxiv.org/abs/2303.13865
English
oh my god I just realized it's called an "equation" because it's got an equals sign that's equating two things
English
Postdoc in Stochastic modeling of shape ensembles - academicpositions.com/share/twitter/… with Aila Särkkä and me
Suomi
Moritz Schauer retweetledi
Applications are open for the CO-OP Master’s 26/27, which is a unique work-integrated programme, combining academic training with industry experience!
Hold a Bachelor’s in math, science or engineering & want to help shape a prosperous Africa?
Apply via: apply.aims-network.org
English
Moritz Schauer retweetledi
Sam Power, Giorgos Vasdekis. [statCO]. Some aspects of robustness in modern Markov Chain Monte Carlo. arxiv.org/abs/2511.21563
English
Moritz Schauer retweetledi
Neural Guided Diffusion Bridges arxiv.org/abs/2502.11909 w/ @gefanyang @MeulenFrank We introduce a new bridge simulation method that combines the guided proposals of @MeulenFrank and @MoritzSchauer with an additional correction drift term parametrized by a learnable neural network. The family of laws on path space induced by these proposals provides a rich variational family for approximating the law of the diffusion bridge. Once the variational approximation has been learned, independent samples can be generated at a cost similar to that of sampling the unconditioned process. The methods is particularly powerful for conditioning on rare events and for simulating multimodal distributions, which pose challenges for score-learning and MCMC-based approaches.
English
Moritz Schauer retweetledi
Sebastiano Grazzi, Giacomo Zanella. [stat.CO]. Parallel computations for Metropolis Markov chains with Picard maps. arxiv.org/abs/2506.09762
Italiano
Right, you don't need error bars on error bars.
Probabilistic uncertainty about uncertainty collapses. This is the “monadic join” in probability.
Instead of a coin with random bias p ∼ π, you can flip a coin with the deterministic bias μ.
Just take μ = E[p].
XKCD Comic@xkcdComic
Error Bars xkcd.com/2110/ m.xkcd.com/2110/
English