Jim Simons was one of greatest philanthropists of our generation in terms of his support of math and science research through the @SimonsFdn. One of the best years of my career so far was spent at the @SimonsInstitute. Thank you Jim. Your legacy lives on.
https://t.co/2g9UH0SVI3
NEW PREPRINT
The harm of the use of imbalance correction methods on calibration, extending our earlier work on regression models to more complex techniques
https://t.co/P1EFWRRn3E
”Solving” imbalanced classes “problem” has been a flourishing cottage industry academic field.
Perusing papers in imbalanced classification has been a real eye opener, one would have thought SMOTE paper with 25K citations is an exception rather than a rule.
Nah. There is a whole cottage industry is out there, a Jurassic Park of solving a problem that is actually not a problem.
Bit too late for that (like 10-15 years late) but if one wanted to become a highly cited researcher in computer science data mining imbalanced classes was the way to go.
Just create some ensemble or yet-another-variant-of-smote, show it works on a few datasets - bingo, you have highly cited paper.
#imbalancedclasses #dataminers
@PaoloCrosetto When statisticians (like me) criticize the poor methodology of so many published papers, people often say "yes but there are not enough statisticians". One of the obvious solutions is to publish less papers. People often chuckle when you say that 🥲. See https://t.co/egdRSUFGn6
Matrix multiplication is simple enough for high school students to grasp yet so complex that even seasoned mathematicians haven’t mastered it. Learn how how AI is advancing this area of math on the Quanta YouTube channel: https://t.co/86VBOv40lY
We received a lot of responses on our recent paper in @NaturePhysics Emergent stability in complex network dynamics
https://t.co/77KykKtwiV
So first of all THANKS. Really exciting to see all the attention.
Now four weeks in, a new thread about the broader picture
Professor Strang gave his last Linear Algebra lecture today after 66 years at MIT.
Strang was among the first to upload his classes to MIT OpenCourseWare when it first came online in the early 2000s. His 18.06 lectures have been viewed millions of times around the world
#OTD 1920 Edward Kaplan b (d 26 Sep 2006) 🇺🇸Best known for the Kaplan–Meier product-limit statistic for estimating survival functions. The 1958 joint paper with Paul Meier is now one of the most cited statistics papers of modern times, although it took >12 yrs to catch on. 1/5
It is time to re-think the scientific publishing model…
Just some issues:
• Open access is too expensive for authors in most cases
• Reviewers and editors provide their time for free
• Process is too slow and limits scientific advancement https://t.co/8OGwLNfcXR
In 1972, Bob Metcalfe presented a thesis about connecting MIT’s mainframe computer to a precursor of the internet called Arpanet. The dissertation committee failed him, saying the topic wasn’t theoretical enough. This year, he received the Turing award. https://t.co/UJnflSjy5S
2023-03-24: CHANGES IN R 4.3.0 SIGNIFICANT USER-VISIBLE CHANGES
*
Calling ‘&&’ or ‘||’ with LHS or (if evaluated) RHS of length greater than one is now always an error, with a report of the form
'length = 4' in coercion to 'logical(1)'
… https://t.co/cOYti0qYoz
Academic hot take: requiring PhD students to publish a dissertation manuscript in a journal before they can graduate is pointless, cruel, and stressful for everyone involved. Please everyone stop doing this.
Breaking 🚨
OpenAI just dropped GPT-4 and it can literally blow your mind 🤯
GPT-4 is a large multimodal model that can accept image and text inputs and emits text outputs.
Capabilities of GPT-4 in a thread 🧵👇