Andrew Delong retweeted
I resigned from Meta, and my internal post got leaked to the press, resulting in some fragmented quotes. Here is the full thing: facebook.com/permalink.php?s…
Andrew Delong retweeted
Super excited by this new #ICML paper: we find that many deep nets have a few (<50) "critical" neurons and we have an efficient alg to find them.
Evaluating these critical neurons enables fast model repair (e.g. reduce bias) and interpretation. Great work by @GhorbaniAmirata!
Andrew Delong retweeted
Replying to @hippopedoid @jhnhw
I occasionally encountered similar artifact with scRNA seq data. The spaghetti-loop in UMAPs changes completely when you shuffle the order of rows in the training data.
Andrew Delong retweeted
The graphics in this article are excellent— this is a huge part of communications that has been missing. If you want to understand how layered interventions reduce #covid19 spread— why masks are indispensable but not always “enough” — read this.
english.elpais.com/society/2…
Andrew Delong retweeted
The 5-minute, low-cost virus test using #CRISPR, a mobile phone, without fancy equipment, for point-of-care
sciencemag.org/news/2020/10/… @RobertFService @ScienceMagazine @NewsfromScience
medrxiv.org/content/10.1101/… @TheOttLab @GladstoneInst @doudna_lab
Andrew Delong retweeted
1/5 Self-Distillation loop (feeding predictions as new target values & retraining) improves test accuracy. But why? We show it induces a regularization that progressively limits # of basis functions used to represent the solution. bit.ly/2HnOACo w/@farajtabar P.Bartlett
Andrew Delong retweeted
Trolley problems ("Kill 1 to save 5?") are good for clarifying moral intuitions, but my grad student Julian DeFreitas argues they're irrelevant to designing driverless cars, which should kill no one. gsas.harvard.edu/news/storie…
Andrew Delong retweeted
#MLCB2020 is happening! A fully independent virtual conference this year, Nov 23-24. Checkout the amazing line up of keynote speakers: mlcb.org/
Abstract deadline: Sept 30th. @anshulkundaje @suinleelab @james_y_zou @david_a_knowles @QuonBio
Andrew Delong retweeted
My biggest regret as a 1st year grad student was pretending to say I knew X when I didn't really. Learning to say "No, tell me about X" became a gateway for (1) learning stuff, (2) getting to know a person, and (3) learning how to say no in other areas of my life.
This is a great idea: construct "separate embedding spaces, each invariant to all but one augmentation." arxiv.org/abs/2008.05659
7
Andrew Delong retweeted
Some news! After many lovely years at @harvardmed I'm moving to @Columbia fall 2020 to start a new lab as an Assistant Professor in Systems Biology and the Program for Mathematical Genomics--and I'm recruiting students and postdocs! Email/DM me or see aqlab.io. 1/2
Andrew Delong retweeted
Interested in using deep learning for designing proteins or cis-regulatory sequences? Try out this new (and fast) method by the very talented @jjohli: arxiv.org/abs/2005.11275
Andrew Delong retweeted
What is your micromort risk of #COVID19?
nytimes.com/2020/05/22/well/…
1 micromort =one-in-a-million chance of dying
by @DRobertsNYC
Andrew Delong retweeted
Join me at 8PM ET tonight for a live @munkdebate Q&A, where we'll discuss #covid19 and the post-pandemic world.
Tune in on Facebook: facebook.com/munkdebates
Or on the Munk Debates website: munkdebates.com/dialogues/sc…
Andrew Delong retweeted
A smart take on virus testing by one of my Theranos sources. You may have heard of him: his first name is Tyler... link.medium.com/sfw4ez9x95
Andrew Delong retweeted
This is a really neat way of using neural networks to get rid of polygonal meshes. The interaction of computer graphics with neural networks is really exciting.
NASA: Neural Articulated Shape Approximation
Representing characters and deformable shapes without polygonal meshes for 3D perception, games, ...
arxiv.org/pdf/1912.03207.pdf
Andrew Delong retweeted
(1/2) New study just out, face masks effective for source control against influenza and seasonal coronaviruses, even reducing virus detection in aerosols in patients with coronavirus infections. 5 years of hard work to generate the data nature.com/articles/s41591-0…
Andrew Delong retweeted
More evidence of pre-symptomatic transmission of COVID19, which I think we must now regard as the reality we are facing. This paper estimates that 44% of transmission could occur 2-3 days before first symptoms. medrxiv.org/content/10.1101/… via @marcelsalathe #quarantineloophole 30/























