Twitter | Search | |
enthusiasm curbed
Chasing the dream of being a race car passenger.
2,153
Tweets
669
Following
65
Followers
Tweets
enthusiasm curbed retweeted
michael_nielsen Sep 13
Fascinating article on the physical demands of chess, and the physical training chess grandmasters do:
Reply Retweet Like
enthusiasm curbed Sep 12
Replying to @LukasSchlogl
Now what if I were to tell you that a significant portion of the global financial industry relies on recent college grads doing “modeling” in Excel...
Reply Retweet Like
enthusiasm curbed retweeted
Tom Stafford Sep 10
"I'm a fiction writer in the grant-proposal genre" (From The New Yorker's David Sipress)
Reply Retweet Like
enthusiasm curbed retweeted
michael_nielsen Sep 11
Someone - I forget who - made the observation that either: (a) we're alone in the Universe; or (b) we're not; and that (c) either is staggering.
Reply Retweet Like
enthusiasm curbed Sep 11
This rings pretty true, and the sad part is that ML is supposedly one of the more practically minded academic fields...
Reply Retweet Like
enthusiasm curbed retweeted
Will Kurt Sep 11
p-values are misused, ultimately, because the people misusing them are not motivated by understanding some underlying process or by science, but by promotion and tenure. No statistical framework will save you from not really caring if you're wrong.
Reply Retweet Like
enthusiasm curbed Sep 11
Replying to @curiouswavefn
Is GEB not long enough for you haha?
Reply Retweet Like
enthusiasm curbed retweeted
John Pfaff Sep 10
This, from a pretty harsh review of Gladwell’s new book, is an EXCELLENT lesson on why you should always track down the original source of seemingly-unlikely statistics. Because often secondary sources launder really bad counting. The review is here:
Reply Retweet Like
enthusiasm curbed retweeted
Lukas Schlögl Sep 6
Just priceless. An estimated 20% of genetic research papers contain errors because Excel converted some gene names into calendar dates.
Reply Retweet Like
enthusiasm curbed retweeted
Jason Sheltzer Sep 11
Our new paper is out today. We used CRISPR to uncover some really striking findings with several drugs and drug targets in clinical trials. Also, we accidentally found the first-ever inhibitor of the cyclin-dependent kinase CDK11.
Reply Retweet Like
enthusiasm curbed retweeted
Jeremy Howard Sep 11
This is the best distillation of recent (and old!) research on batchnorm I've seen. There is so much to learn about training mechanics by studying this thread and the links it contains.
Reply Retweet Like
enthusiasm curbed Sep 8
This is great! Worth posting on
Reply Retweet Like
enthusiasm curbed retweeted
Farzad Mostashari Sep 7
1/ tweetstorm coming Warning: strong opinions The analysis is naive and the findings are ridiculous; the fact that it was published is a sign that when medical journal editors hear "deep learning AI" their brains stop working.
Reply Retweet Like
enthusiasm curbed Sep 7
Replying to @citnaj
“a lie is halfway round the world before the truth has got its boots on”
Reply Retweet Like
enthusiasm curbed retweeted
Sean J. Taylor Sep 5
This is a pretty short and unpolished <thread> on launch criteria for experiments. Hoping for feedback! Background: one heuristic people use to decide to "ship" in an A/B test setting is p-value < 0.05 (or maybe 0.01). How important is "stat sig" for maximizing expected value?
Reply Retweet Like
enthusiasm curbed retweeted
Arvind Narayanan Sep 1
This is a thread about the most interesting subject I learned recently. It started when I I read the paper “How Complex Systems Fail”. Like many others encountering it, I felt it was the most insightful four pages ever written about software engineering
Reply Retweet Like
enthusiasm curbed retweeted
Rachel Thomas Sep 4
Essay grading software (used in 21 states) focuses on metrics like sentence length, vocab, spelling, & subject-verb agreement, but ignores hard-to-measure aspects like creativity. Meaningless gibberish essays created with sophisticated words score well.
Reply Retweet Like
enthusiasm curbed retweeted
Hillel @ strangeloop Sep 4
The other day I was talking with a friend in public policy. She was worried about how decisionmakers hide behind private, unaudited software. I think this is a bigger problem than even programmers suspect. "Science code" is, almost by definition, broken.
Reply Retweet Like
enthusiasm curbed Sep 5
Replying to @gappy3000 @Meaningness
This was great, reminded me of wonderful read by
Reply Retweet Like
enthusiasm curbed retweeted
gappy is next level Sep 4
How do people learn to cook a poisonous plant safely?
Reply Retweet Like