Fred Rahmanian ScD

Today in Nature, we show how a standard neural net, optimized for compositional skills, can mimic human systematic generalization (SG) in a head-to-head comparison. This is the capstone of a 5 year effort with Marco Baroni to make progress on SG. (1/8) nature.com/articles/s4158…

Starting off the morning at @MGHCancerCenter #CancerEquityColloquium with Dr. Karen Winkfield “Improving Cancer Equity Through Precision Medicine”. Critically important point - it’s not just about the molecule.

The ‘0 / 1 / Done’ Strategy for Data Science by Marios Perrakis link.medium.com/8PI2aePj0ob

Transformer models: an introduction and catalog by @xamat link.medium.com/FJIgFiN9Wob

Shameful.

@Grady_Booch Wishing you the best outcome.

Your doctor sends you for a lab test. Ever wondered if the outside temperature on the day of the test changes its result? Neither did @devingpope or I… until we did this study! Turns out, temperature distorts results Thread—new paper in @MedCellPress twitter.com/CellPressNews/…

When you think you're good at commenting your code but then see some of Apple's open source stuff

I just wrote a post about some of the work we're doing at @SeerMedical. We're going to be using the blog to share some of our work in design & engineering for neuroscience and epilepsy. medium.com/seer-medical/t… RT's much much appreciated!

@rasbt Nice. But unfortunately still no support for M1 class chips in sight.

Re what can possibly be added and improved, this is really impressive: PyTorch 1.10 --> "This release is composed of over 3,400 commits since 1.9, made by 426 contributors." PS: more detailed parameterize tutorial here: pytorch.org/tutorials/inte…

I always think that PyTorch is already so feature-rich and polished, what could they change and/or add? A neat additions is the "parameterize" module. Below, a quick example creating a custom layer. However, the real cool use-cases are applying it to larger modules of course!

Not much of a Saban fan but this is great.

Continuing the discussion, I just saw the "Deep Neural Networks and Tabular Data: A Survey" paper uploaded to arxiv the other day: arxiv.org/abs/2110.01889. Useful list of most (/all?) the approaches for classification (left) and data generation (right) so far

Hi Twitter. I need your help. A few years ago, I had a freak accident that sent me into catastrophic kidney failure. I've been struggling to find a kidney donor ever since. I'm in dire need of a type O+/- kidney and ask that you please share my story. #kidneyfordevon 🧵👇

Deep learning seems unstoppable! I'm particularly impressed by the recent progress of deep learning on tabular data. This new survey paper provides an overview of the SOTA deep learning methods on tabular data. A great read for students and practitioners. arxiv.org/abs/2110.01889

"Aristotle famously talks about courage being on a spectrum, cowardice on one end, recklessness on the other, the desire to engage in a decision has to rest at a midpoint between these extremes." @RyanHoliday in his great conversation with @RyanHawk12 learningleader.com/ryanholiday437/

@Saints @BenjaminSWatson @demario__davis @RobinRoberts Love Ben and love Louisiana.

👏 @BenjaminSWatson @demario__davis @RobinRoberts

@gnayyar Ahem. No masks?

Missed it? Can catch the insights on-demand through Salesforce+ ➡️ salesforce.com/dreamforce/

Can't get last week's @Dreamforce off my mind! 🎉 I was inspired and energized by the leadership presence at #DF21 and loved meeting some of my @salesforce Dream Team colleagues in person! It was great to safely celebrate being together and enjoy the moment.

📊🧵 Collection of tweets featuring open-access materials that I have shared over the last years: Talks, seminars, blog posts, hands-on notebooks, codes, and more! #rstats #ggplot2 #tidyverse #dataviz 🧙‍♂️

Like this. I look forward to part 2.

PLEASE RETWEET: an opportunity for Stat/Biostat PhD applicants from historically marginalized groups to get support on their PhD applications. Apply now!! @AmstatNews @InstMathStat @UWStat @UWBiostat