Someone has made a folder named `~` on our server... This will not end well...
— Thomas Lin Pedersen (@thomasp85) August 23, 2018
Someone has made a folder named `~` on our server... This will not end well...
— Thomas Lin Pedersen (@thomasp85) August 23, 2018
How to tell if you’re arguing with a bot https://t.co/YEaE5Q3H8A
— MIT Technology Review (@techreview) August 22, 2018
People usually get information from others in a multi-turn conversation. To approach this, we’ve released CoQA 🍃—A Conversational Question Answering Challenge by @sivareddyg•@danqi_chen•@chrmanning. 127K Qs— free-form answers—with evidence—multi-domain. https://t.co/XkdWC2Hh3W pic.twitter.com/v0FXMnNg8n
— Stanford NLP Group (@stanfordnlp) August 22, 2018
Prime example of why all clinical trials should be published regardless of results.
— Chris Chambers (@chrisdc77) August 22, 2018
And the best way to do that, as the authors note, is to decide what gets published before results are known. Common sense.
This option exists. It's real. Let's use it. https://t.co/VKgcVWgO3m https://t.co/IKs4BNbB0B
✨ Neat text prediction app w/ #tidytext by Nate Day
— Mara Averick (@dataandme) August 21, 2018
🔮 "Text Prediction: Predicts next word based on previously parsed n-gram look up" https://t.co/OY6a7wjV1T #rstats #rshiny #NLProc pic.twitter.com/tsSuo1T6nX
I updated my guide with new GPU recommendations: RTX 2080 most cost-efficient choice. GTX 1080/1070 (+Ti) cards remain very good choices, especially as prices drop. Some discussion on TPUs/AWS — can be good in some cases. https://t.co/EVTJ7UnQFt pic.twitter.com/bHCuAsh0uw
— Tim Dettmers (@Tim_Dettmers) August 21, 2018
To you, this image looks like random shapes. But to AI-powered online filters, it's 96.7% sure it's "explicit nudity." Here's my piece on what AI artwork like this can teach us about the algorithmic gaze https://t.co/H4siR32Alf pic.twitter.com/bqp3asciZP
— James Vincent (@jjvincent) August 21, 2018
Well I'm screwed https://t.co/2YHdV0tjG0
— Jeremy Howard (@jeremyphoward) August 21, 2018
Some "oldies but goodies" from my blog:
— Data Science Renee (@BecomingDataSci) August 20, 2018
-How to use Twitter to learn data sciencehttps://t.co/20rmG4aaRU
-A Challenge to Data Scientistshttps://t.co/d3cqj1x8Ag
-Boosting as a metaphor for diverse teamshttps://t.co/PUS9vNfYrP
-Intro ML terms w/foodhttps://t.co/oUx6wqUlZM
This insight from @lessin explains a lot about Trump: "the only possible future of privacy was to generate lots of disinformation rather than try to prevent information from being exposed in the first place. In other words, bury the truth in a sea of lies"https://t.co/54SFxaO0ec
— Tim O'Reilly (@timoreilly) August 20, 2018
LGBTQ women and women of color were more likely than their straight, white counterparts to have been harassed, and women of color were more likely to report feeling unsafe because of their gender. 4/4 pic.twitter.com/nB6ZPfchLS
— Rachel Thomas (@math_rachel) August 20, 2018
“Scientists have equated rigor and being critical with being cruel." 3/4 pic.twitter.com/ltZQ2722aO
— Rachel Thomas (@math_rachel) August 20, 2018