Bliss

Tables are about to get a whole lotta love in patchwork #rstats

Apple say their new sleep apnea prediction tool for Apple Watch is "validated in a clinical study — unprecedented in size for sleep apnea technology" - yet no reference provided Concerning, as 'validated' does not equal valid Can anyone source the study #AppleWatch10

I'm excited to share the next version of patchwork with the world, available on CRAN now. First class support for gt tables and even more freedom with `free()` tidyverse.org/blog/2024/09/p…

Coming soon to a bookstore near you! Epidemiology: an introduction KJ Rothman, KF Huybrechts, EJ Murray 🥳🥳

I trained a LLM for single cell RNAseq generation recently. Terabytes of data. Megawatts of GPU usage. Lights in houses dimmed all throughout silicon valley. In the end, it learned to predict cell type. As in, the predictions were only slightly better than the mean expression!

Interesting work on reviving RNNs. arxiv.org/abs/2410.01201 -- in general the fact that there are many recent architectures coming from different directions that roughly match Transformers is proof that architectures aren't fundamentally important in the curve-fitting paradigm

This is turning into a cringe fiesta

The Notorious MBJ takes open science to a new level my making it a living model. New data (RCT) will continually update the posterior estimates. *whistles a happy Bayesian jingle*

Amazing resource from Nicola. 🌟 The course website is here: nrennie.rbind.io/training-bette… (including slides)

Some news: I've just open-sourced the draft of a book I'm working on about how #rstats tidymodels users can make their code run faster without sacrificing predictive performance! simonpcouch.com/blog/2024-10-2…

Remember Golden Gate Claude? Etowah Adams and I have been working on applying the same mechanistic interpretability techniques to protein language models. We found lots of features and they’re... pretty weird? 🧵

google pair explorables has some cool blogs and articles on foundational ml and mechanistic interpretability

Great to see this R package: stephenrho.github.io/pminternal/ 😀 Implements the (in)stability plots and checks recommended by myself & Gary Collins after developing a prediction model Our paper is here: onlinelibrary.wiley.com/doi/full/10.10…

[#Video] Our latest QuanTIM webinar's video is online ! ⤵️ 💬 Machine learning for health: promises and methodological challenges 🗣️ Gael Varoquaux 🦋 (soda_INRIA @[email protected],Centre Inria de Saclay) 💻 sesstim.univ-amu.fr/fr/video-box/w…

New version 0.2.0 of tidyplots.org now on CRAN 🚀 🐞 Important bug fix of standard deviation plots 🎄 More customization of fonts and themes 🤖 New heuristic to determine dodge_width #rstats #dataviz #phd

Wrangling string columns for machine learning, the new StringEncoder in skrub gives such a good compute/prediction performance tradeoff. It's mostly just a bunch of simple tricks, but with well-chosen defaults. This is what we love in skrub

The report is a thorough compilation of many expertise into a nuanced view of various aspects of AI risks, from the heavily publicized extreme risks to social consequences, bias or resource consumption. Proud to have been part of this effort. Thanks Yoshua Bengio for leading it

I‘m excited to announce that we raised €9m in a pre-seed round for Prior Labs, the world’s first company focussed on building tabular foundation models! The round was led by Balderton Capital, with participation from XTX Markets, Hector Foundation, Atlantic Labs, and Galion.exe. We’re