“It’s (already) in the weights.”-Peter Steinberger. #DataSkyence
Latest posts tagged with #Dataskyence on Bluesky
“It’s (already) in the weights.”-Peter Steinberger. #DataSkyence
Learn how to:
✔️ Navigate #CommonFund data resources
✔️ Select and implement multiomics analysis methods
✔️ Integrate datasets across multiple data types
✔️ Apply workflows that support reproducible research
✔️ Manage conflict within research teams
Apply now: www.jax.org/education-an...
#DataSkyence
Portrait of George Boole with a quote: "No matter how correct a mathematical theorem may appear to be, one ought never to be satisfied that there was not something imperfect about it until it also gives the impression of being beautiful."
George Boole, who died #OnThisDay in 1864, was an English mathematician whose work laid the foundation for modern computing. Boole developed Boolean algebra, a system of logic that underpins digital circuits and computer science, making today’s digital world possible.
#dataskyence #ComputerSkyence
The DoubleML rhino shouting out "User Survey", available via https://forms.gle/HjZsWgrF5UEFUyDX9
📣 Announcing the First DoubleML User Survey! 📊
We’re excited to launch the first DoubleML User Survey! We’d love to hear from all prospective, new, or experienced users of DoubleML in Python or R.
🔗Please take part here: forms.gle/HjZsWgrF5UEF...
#EconSky #CausalSky #dataSkyence #Python #R
Need some good #datascience #churn brainfood for the holiday weekend? There's a neat new research paper in Nature Magazine ScientificReports on using Extreme Learning nets to predict #churn! #dataskyence www.nature.com/articles/s41...
👩🏫 💻 STEM educators: How can your students gain experience with real-world #genomics data?
The Database Tutorials module provides guidance for using #BLAST, #cBioPortal, #PolyPhen-2 and #UniProt in classrooms or independent projects: www.jax.org/education-an...
#Dataskyence #EduSky
Flyer for Biomedical Data Science in Context program. See website for more information at www.jax.org/bdsic.
🖥️🧬 Through a combination of technical training, mentorship and career planning, you’ll come away with:
✅ A solid foundation in biomedical data science
✅ Confidence in your computational thinking
✅ A personalized plan for your next academic or career move
#DataSkyence
Last day to register for our BENA Skills Camp in September in Berlin! #EconSky #EconConf #dataSkyence #CausalSky
How insane is it that openai asks for BIOMETRIC information for using its API with the o3 model?! Asking me to take a selfie and upload personal documents. Anthropic, here I come #dataskyence
White graphic with blue along the bottom reads, ""The afternoon workshops offered extensive hands-on experience, allowing me to seamlessly incorporate these skills into my existing workflows and behavioral analysis pipelines." Tushar Arora | Postdoctoral Fellow | Albert Einstein College of Medicine | The Jackson Laboratory."
Join @jacksonlab.bsky.social's #MachineLearning course for hands-on training & expert lectures on:
🔬 Experimental workflows
📈 Analysis tools
🐁 Animal tracking algorithms
📊 Statistical modeling
Build skills in quantitative #BehSci analysis & #DataSkyence: www.jax.org/education-an...
#Neuroskyence
If I never do speaker annotation again it'll be too soon #dataSkyence
💡 Research at the intersection of data science, machine learning and behavior analysis?
This @jacksonlab.bsky.social Short Course offers expert lectures and hands-on training to help you apply #MachineLearning methods to your research studies: www.jax.org/education-an...
#Dataskyence #Neuroskyence
#datascience #dataskyence #art #artist #literature
Communities, please have talks and read up about AI and AI related abuses and copy-right problems, then fill out the EU consultation, hopefully we can limit the damage AI will cause in the future with this! ec.europa.eu/eusurvey/run...
left panel :"don't make me tap the sign" right panel: "cite software packages in your methods section"
I would save time by leaving this comment in my peer review template instead of adding it in every time
#AcademicSky #PsychSkiSky #DataSkyence #rstats
Screenshot of an interactive world map using the Robinson projection showing 2023 aid commitments in each country
Western Europe mapped with a cropped Robinson projection with labels for each country
Egypt mapped with a cropped Robinson projection with labels for each country
Table of contents for the post: Working with map data Get map data Maps and projections with Observable Plot Built-in projections Other projections Filtering map data and adjusting projections Removing elements Quick and dirty cheating method: change the width or height Built-in projections and domain settings Other projections and .fitExtent() Arbitrary areas and .fitExtent() Working with USAID data Get USAID data Connect USAID data to the map data Improving the map Fixing labelling issues Some final tweaks The full game: Complete final code
New blog post! After archiving USAID data last week, I decided to play around with and recreate parts of the ForeignAssistance dot gov dashboard with #QuartoPub and #ObservableJS. I learned a ton about D3 projections and making pretty Javascript plots #dataskyence www.andrewheiss.com/blog/2025/02...
If you have thoughts on technical requirements on data collection and experimental design to improve microbial machine learning, @asterainstitute.bsky.social wants to hear from you
substack.com/@manjarinara...
#StatsSky 🧬🖥️ #Dataskyence
Got the gem in the mail the other day and dove right in rather than post about it.
One chapter in and I have to say: It is worth every penny! The combination of graphics, explanations and code is valuable to those of us experienced with LLMs and those not so much!
#dataSkyence #booksky
Orange graphic with an illustrated person wearing a white Jackson Laboratory lab coat working on a laptop and the words, "FAIR data" across the top left of the graphic.
📊 What does it mean for data to be FAIR?
Discover the four guiding #FAIRdata principles in less than 15 minutes and learn why you should apply them to your biomedical research in @jacksonlab.bsky.social's online course: education.learning.jax.org/fair-data-pr...
#DataSkyence
Plot showing respondent demographic proportions compared to national proportions, overlaid on variable-specific regions of practical equivalence (ROPEs)
Table showing sample characteristics compared to nationally representative Current Population Survey (CPS) estimates, with probability that the sample is equivalent to the national proportion
Also, it’s important to check the variable details to check for availability. While basic demographic variables like age, sex, marital status, etc. are available in both the monthly surveys and in the annual ASEC, more specialized variables are not. Variables related to philanthropy and volunteering are only available in September (since they’re part of a special CPS Volunteer Supplement), and only in some years: [Screenshot from the IPUMS website showing the availability of the volunteer status CPS variable]
Contents for the post Nationally representative demographic data Accessing US Census data ACS CPS (and others!) Getting started Getting CPS data from the IPUMS website Finding variables Selecting samples Downloading the data More reproducible alternative: using the IPUMS API Loading CPS data Summarizing CPS data Weighting Calculating population-level proportions Summarizing sample proportions Testing sample vs. population proportions frequentist-ly One-sample proportion test for age One-sample proportion test for volunteering Proportion tests and differences for everything all at once Testing sample vs. population proportions Bayesian-ly ew null hypothesis significance testing Modeling proportions with a binomial distribution Working with the posterior The region of practical equivalence (ROPE) Bayesian proportion test for volunteering Posterior proportions, differences, and ROPEs for everything all at once
New blog post! Here's an #rstats guide for how to (1) get CPS data from IPUMS and (2) compare sample and population proportions both frequentistly and Bayesianly with {brms} (with ROPEs!), and (3) make pretty plots and tables #econsky #polisky #dataskyence www.andrewheiss.com/blog/2025/01...
These two books, by @anilananth.bsky.social and @tomchivers.bsky.social, are the first two books in a very long time that I read in their entirety without significant pause or other diversion along the way.
I cannot recommend them enough!
#booksky #dataSkyence
The causal part:
If you are a #causal #DAG enthusiast, you'll finde some causal diagrams and a discussion on causal aspects of demand analysis in the paper too 😀
#CausalSky #dataSkyence
🧪 In your field, tell me about an experiment or empirical paper broadly construed you find beautiful!
Inspired by a paper from the #PhilSci feed. #neuroskyence #dataskyence #MLSky #StatSky 🧬💻
My highlights from Axios AI+SF yesterday:
1. EarthSpecies.org - AI to communicate with animals! ❤️
2. Databricks CEO Ali Ghodsi says #AI "definitely a bubble" after receiving a $62B valuation in Series J 🤯
3. Waymo claims #driverlesscars are already safer than human drivers 🤔
#dataskyence
(1) This blog post captures why I generally don't like multi-verse analyses.
- They take no position on what constitutes valid and invalid analyses; and
- All the analyses don't even correspond to the learning the same estimand for a given task.
#MetaSky #StatSky #dataskyence
econometrics question
is there a specific argument why the residuals ought to have a mean of zero in single-variable time series models at all? (given they aren't distributed normally, which can just be transformed)
#dataskyence #econsky #econometrics
Google says that #LLMs have plateaued and need new breakthroughs - just don't call it "hitting a wall" 😉 #ai #genAI #aihype #dataskyence #neuroai www.businessinsider.com/google-ceo-s...
𝐅𝐈𝐍𝐀𝐋 𝐂𝐀𝐋𝐋 𝐟𝐨𝐫 𝐒𝐔𝐁𝐌𝐈𝐒𝐒𝐈𝐎𝐍𝐒
Data Transparency Special Issue call is closing on Friday 13th December
Do get in touch if you would like to submit, but may miss the deadline - contact@ijpds.org
ijpds.org/calls/data-t...
#DataTransparency #OpenCall #EpiSky #PublicHealth #Sociology #dataSkyence
The chain of inference in mediation is pernicious and fraught with peril. A similar fallacy occurs in basically all research of biomarkers as intermediate clinical endpoints and most of biomedical science is asleep on this front.
#neuroskyence #CausalSky #Medsky #dataskyence
here's what i spent today making for fun: Victoria's year 9 average NAPLAN score of a high school versus its suburb's median house price, with some outliers and extrema labelled
disclaimers below
#NAPLAN #dataSKYence #Econsky