This seems like a big deal. https://t.co/KXy2CMMopk
— Hilary Parker (@hspter) September 20, 2018
This seems like a big deal. https://t.co/KXy2CMMopk
— Hilary Parker (@hspter) September 20, 2018
just your friendly reminder that machine learning is sometimes another term for "over fitting at scale"... https://t.co/axu6sjcSbn
— JD Long (@CMastication) September 20, 2018
training to overfit a new model on a small dataset (or existing model on new data) is prob the most important and useful rule in practical DL
— helena sarin (@glagolista) September 20, 2018
(just found a bug in my new code for synthetic data generator 😌)
“Help! I can’t reproduce a machine learning project!” - nice blog post from @rctatman https://t.co/eTCEmW2Qao
— Ben Hamner (@benhamner) September 19, 2018
God grant me the serenity to accept the statistical tools I cannot change, courage to adopt new languages & methods when I can, and wisdom to not debate SAS vs. Python/R on LinkedIn/Quora. #pydata #rstats #jupyter
— JD Long (@CMastication) September 18, 2018
Things you see on the internet https://t.co/LcOCY48NM1
— hardmaru (@hardmaru) September 17, 2018
IBM is being sued for age discrimination. A class action. https://t.co/bkDslDvYqi
— Ariana Tobin (@Ariana_Tobin) September 17, 2018
It's a good day to reread our investigation from March. 1,000+ ex-IBM workers helped @PeterGosselin and I figure out why so many older workers had lost their jobs: https://t.co/ACGiwPMjsa
Information abstraction is vital for progress. It's also why most of us have imposter syndrome. No-one understands everything. We try, then test, then trust, the information abstractions we use. Lean in to that. It's why thoughtlessly flicking a switch lets you light a room.
— Smerity (@Smerity) September 17, 2018
For anyone debating the value of using physical simulations to improve forecasts, here's Hurricane Florence's actual track compared to all other tropical systems in the area in historic Septembers
— Ben Hamner (@benhamner) September 17, 2018
A simulation-based approach nailed the forecast track. Pure ML would fare terribly pic.twitter.com/QSaNzaLGMc
In which I calculate the Gini coefficient of various things. I had fun. https://t.co/yIlD3LI8xs
— Tim Harford (@TimHarford) September 17, 2018
Statisticians are changing the way we measure poverty https://t.co/zdRVt17mRj by @HetanShah @SocialMetricsComm
— Tim Harford (@TimHarford) September 17, 2018
Animated GIF Loops produced by students in @Golan’s course on Interactivity and Computation. https://t.co/WIkiVIGxk6 pic.twitter.com/HaK6ZL1chm
— hardmaru (@hardmaru) September 17, 2018