I just finished an adventurous visit to Gelman's blog https://t.co/Cz5DRb0oPy
— Judea Pearl (@yudapearl) June 21, 2019
and I am sharing here the last sentence: "if Causal Inference is 'Statistical Inference given causal assumptions' [as some claim] then Car making is car-painting given an engine and a body."#Bookofwhy
See @FT's use of @googleearth studio to visualize the progression of the Hong Kong protests against an extradition bill, via @sdbernard's tweet https://t.co/TxRiNd9SXi. More in Top 10 #ddj: https://t.co/Hhvh1yuWqs @nicolle_liu @suelinwong @alicemuwu @AdrienneKlasa @charlesmok pic.twitter.com/8q0bwYHBwX
— GIJN (@gijn) June 21, 2019
Let's clarify this...
— François Chollet (@fchollet) June 20, 2019
On one hand, the supermajority of SF/SV founders were born rich, went to an Ivy (nearly a requirement to get funded), are part of the network.
On the other hand, SF/SV rhetoric is all about rag-to-riches stories, lionizing founders as self-made heroes
This AI was able to correctly identify manipulated images 99% of the time, compared to a 53% success rate for humans. https://t.co/OeodgJiz8H
— MIT Technology Review (@techreview) June 20, 2019
Hyperparameter tuning magic pic.twitter.com/t7X1FLdjXQ
— François Chollet (@fchollet) June 20, 2019
NEW: YouTube is under investigation by the FTC, and the probe is in late stages, for its handling of kids and videos for them. This comes as company considers big changes to its platform. Me + @lizzadwoskin + @craigtimberg https://t.co/EW18vwQ21I
— Tony Romm (@TonyRomm) June 19, 2019
I'd like to see a privacy policy written in the style of William Faulkner https://t.co/VhlLEAdJc7
— Peter Skomoroch (@peteskomoroch) June 18, 2019
This is an AWESOME use for machine learning: “defragging” calendars to give back blocks of uninterrupted time! Brilliant. I want this. https://t.co/qvJG1XFBAK
— Tim O'Reilly (@timoreilly) June 17, 2019
I recently got interviewed for a feature on ups and down of being a Data Scientist. https://t.co/Fdt43atuAD#datascience #ml #ai
— Bojan Tunguz (@tunguz) June 17, 2019
cc. @_ColinFay @lebigot @TesseractAcade1
"... I considered teaching a necessary evil that took me away from the lab bench ... But when my life took an unexpected turn, I realized how wrong I had been:" A retired professor reflects on gaining a love for teaching in this week's #SciMagWorkingLife. https://t.co/od6lcUO9Fq
— Science Magazine (@sciencemagazine) June 17, 2019
PS. This is probably why you should call it "empirical journalism" rather than "data journalism". It's not about using data for data's sake. It's about figuring out what works. Knowing (& telling readers) when "the data" *isn't* meaningful or reliable is essential to the mission.
— Nate Silver (@NateSilver538) June 17, 2019
“Data alone isn’t insight. […] The key is to listen to what data has to say — and develop the openness and interpretive skills to understand what it is telling us.”
— Angela Bassa (@AngeBassa) June 16, 2019
This, from @Neil_Irwin on ‘data survivorship bias’ and ‘digital exhausts,’ is very good.https://t.co/X9ITMjZAv8 pic.twitter.com/FAR5hg5OPa
