Modern economies are certainly good at a certain kind of resource allocation, but they're crap at allocating the most valuable resource -- people's potential.
— François Chollet (@fchollet) February 17, 2019
To start with, having to earn $$ is such an enormous distraction from everything you *could* be doing.
You know it.
This figure is striking.
— John B. Holbein (@JohnHolbein1) February 16, 2019
It shows the relationship between where people think they are in the income distribution and where they actually are.
Observations below the line indicate that people think they are poorer than they really are (and the reverse above the line). pic.twitter.com/exECjYOdaH
completely agree: remarkable ≠ semantically coherent. system has no idea that a unicorn has one horn, blithely asserts it has four, etc. simultaneously close and not close to Searle’s room, not at all close to genuine understanding. https://t.co/eVpDUsi6B3
— Gary Marcus (@GaryMarcus) February 16, 2019
always blows my mind how wealth inequality certainly appears to be directly, inversely proportional to top marginal tax rate pic.twitter.com/DjdmPz3ZLQ
— Christopher Mims 🎆 (@mims) February 15, 2019
Terrific career advice here!
— jacobian (@jacobian) February 15, 2019
I'd add that the three step plan she lays out (learn SQL, learn a language, learn to work in the cloud) is a terrific baseline for ANY software eng role. I'd hire someone with those general skills over more limited specific ones in nearly every sitn. https://t.co/G5rJVjU723
I’m not sure I’m as pessimistic as Vicki overall but
— Julia Silge (@juliasilge) February 15, 2019
📌I have absolutely seen what she points out about ML engineer jobs vs. data science jobs
📌Her practical advice about what to do is 🔥🔥🔥 https://t.co/WuZ1Pez2PI
What started off as “oh nice! Impressive results” regarding today’s OpenAI announcements has turned into deep cynicism and mistrust because of their way of handling this matter. I think it is a research win and policy failure.
— Delip Rao (@deliprao) February 15, 2019
Related: https://t.co/SJhgc1RQYw
— Rachel Thomas (@math_rachel) February 15, 2019
It's great to hear from informed and thoughtful reporters like @jjvincent, who are working to put AI advances in context, and consider their wider implications.
— Jeremy Howard (@jeremyphoward) February 15, 2019
His latest piece on @OpenAI's new language model is very nicely done.https://t.co/HduGSutCLL pic.twitter.com/0g7X3Gh9vc
In terms of the underlying techniques, little has changed regarding the underlying model. The lovely thing about deep learning is that a slightly tweaked problem at a new scale can reveal an entirely new result. It's the "scale it until it breaks (or we do)" line of research ;)
— Smerity (@Smerity) February 14, 2019
3 reasons I recommend learning math on an as-needed basis (as opposed to trying to complete a bunch of math pre-reqs before you start on the thing you care about):https://t.co/4pXm4lXmRj pic.twitter.com/8FN76IAIm9
— Rachel Thomas (@math_rachel) February 14, 2019
A personal❤️ in GPT2 paper: the discussion on language-modeling & multi-tasking. We still see papers being rejected bc LM is not considered "useful"(ex Transfo-XL)😢 If your dataset is diverse enough LM is a proxy for a *huge* multi-task objective over pretty much every NLP task! https://t.co/MeJJVrXdIg
— Thomas Wolf (@Thom_Wolf) February 14, 2019
