Help! What precisely is "inductive bias"? Some ML researchers are in the opinion that the machine learning category of ‘inductive biases’ can allow us to build a causal understanding of the world. My Ladder of Causation says: "This is mathematically impossible". Who is right? 1/
— Judea Pearl (@yudapearl) February 14, 2021
💵 Did you know that there is a new open-source @Stripe extension for @Code? It makes it easy to monitor API request logs in real-time, generate sample code, trigger and forward webhook traffic, debug configs, and more.
— 👩💻 Paige Bailey @ 127.0.0.1 🏡 #BLM (@DynamicWebPaige) February 13, 2021
📄: https://t.co/CyAYjkqAnS
💻: https://t.co/N8rEwr63S5 pic.twitter.com/l4LIiqdJAr
I have worked on this problem for a long time at this point. Here is the best strategy I've come up with:
— Chris Albon (@chrisalbon) February 13, 2021
1. Remove educational requirements in job postings. No "CS degree or similar" etc.
2. Read the cover letter, let folks explain why they are right for the role.
1/n https://t.co/Lnee1f0SqH
Now that it's archived, it really seems a shame that the great work that went into Swift for Tensorflow didn't go into Julia instead: https://t.co/APzeNEXuzz
— John Myles White (@johnmyleswhite) February 12, 2021
One of my favorite VSCode features is the regex search/replace that highlights matches in real-time.
— Jake VanderPlas (@jakevdp) February 12, 2021
It's the first time for me that regular expressions have felt intuitive to use. pic.twitter.com/a5ki5cGFnq
Regarding generalization in neural nets, one of the "Things that don't matter very much: The optimizer"
— Sebastian Raschka (@rasbt) February 12, 2021
(via Tom Goldstein's talk, "An empirical look at generalization in neural nets" https://t.co/CqLeHLte3H) pic.twitter.com/uzrvDCirrl
Attention in Dex compared to Einsum. As requested by @_rockt .
— Sasha Rush (@srush_nlp) February 12, 2021
Key differences -> abstracted dimensions, semantic names, non-linear functions, no repeated renaming.
(Two versions in Dex one verbose, one with inference.) pic.twitter.com/V9FcKJMoVq
When you are rich you value safety and sunshine. Map shows the international migration patterns of high net worth individuals. Australia is the place to be! Source: https://t.co/y8H0OPeYQb pic.twitter.com/I4EWCWZsnb
— Simon Kuestenmacher (@simongerman600) February 12, 2021
Several Asian and European nations share the fate of South Korea's shrinking population. Japan has been shrinking for a while, China will start shrinking by the late 2020s. Economies and societies need to be restructured in light of such changes. Link: https://t.co/upE3UtnvQQ pic.twitter.com/w4e5fUWj1Y
— Simon Kuestenmacher (@simongerman600) February 12, 2021
High-Performance Large-Scale Image Recognition Without Normalization
— AK (@ak92501) February 12, 2021
pdf: https://t.co/THe2NfRI1K
abs: https://t.co/Z68FevANZP
github: https://t.co/Gvw5s5HZIh pic.twitter.com/PGrLhn5oyl
I’m excited to share a new textbook @mrtz and I wrote: "Patterns, Predictions, and Actions: A Story about Machine Learning." 1/5https://t.co/3gCH2afaYM
— Ben Recht (@beenwrekt) February 11, 2021
Rather than fall for the bait of certain prominent members of the ML community who are known for their trollish remarks anyway, I'd just like to remind everyone that DeepMind continues to put out tons of impactful research *and* gives employees generous European-length holidays.
— David Pfau (@pfau) February 11, 2021
