From now on this site uses only one AMP-compliant theme to serve all requests. The reading experience should be a lot better on mobile or smaller screens. There are still some small details that need tweaking and maybe some undiscovered bugs, though.
Future of Machine Learining is Tiny
Why the Future of Machine Learning is Tiny: https://t.co/cvwsFciN5m
— Pete Warden (@petewarden) June 11, 2018
Nice write-up by @petewarden on running ML on tiny computers. It's now possible to run a MobileNetV2 vision model at 1 frame per second for a year on a coin-sized battery on a low-end DSP. https://t.co/zx2zZC2jZc
— Jeff Dean (@JeffDean) June 11, 2018
Twitter Data Collection
(Long thread. Click on the tweet to read the full thread.)
I've now reviewed a few papers that use Twitter data but not rtweet😢. I try not to be the reviewer who demands everyone use/cite my work. Instead, I spend my energy making sure data collection is clearly, accurately, & properly described. At a minimum, I think this means:<list>
— Mike Kearney📊 (@kearneymw) June 10, 2018
ACL 2018 Best Papers
Our warmest congratulations to the authors of the ACL 2018 best papers: https://t.co/yXEs95vUp3 #ACL2018 #NLProc @acl2018
— ACL2018 (@acl2018) June 10, 2018
Notable Research
Great work by Yanan Sui on Safe Bayesian Optimization applied to real human clinical experiments!https://t.co/DVL4QxKbjM
— Yisong Yue (@yisongyue) June 9, 2018
Appearing at @icmlconf #icml2018 pic.twitter.com/MAZhvFZRQc
The code for our recent work, "Imitating Latent Policies from Observation," is now on github! @him_sahni @yscrcrhttps://t.co/u2xeA5rJP5
— Ashley Edwards (@RealAshEdwards) June 11, 2018
"A Simple Method for Commonsense Reasoning," Trinh and Le: https://t.co/NnfE3HgwZb
— Miles Brundage (@Miles_Brundage) June 11, 2018
Language models used to improve performance on Winograd Schemas, etc. (still far short of human performance)
Tutorials and Resources
💫 infosec code-through:
— Mara Averick (@dataandme) June 10, 2018
"Anomaly Detection and Threat Hunting w/ Anomalize" ✏ @holisticinfosechttps://t.co/CwRohZ3BTj via @bizScienc #rstats #infosec pic.twitter.com/Gwmy2lXvLA
New book chapter: Optimize the data–ink ratio. #datavizhttps://t.co/ULy0vgOJzH pic.twitter.com/eeC6phXl9h
— Claus Wilke (@ClausWilke) June 10, 2018
Miscellaneous
Michael Jordan warns us that e.g. deep learning is going to lead to "... bad medical decisions being made for hundreds of thousands of people per day ... we're not even close to being able to solve those problems and we're acting as if we are." https://t.co/mXMnIYZGpC
— Max Little (@MaxALittle) April 22, 2018
HBS: Hiring Your First Chief AI Officer - @AndrewYNg https://t.co/yk6E51U3nV
— Peter Skomoroch (@peteskomoroch) June 11, 2018
Github and Open-source
"Github and Open-source Is a Boon for the Underprivileged"
— Amjad Masad (@amasad) June 10, 2018
I typically try to stay away from debating hot topics because it takes time to form an informed opinion but this was a subject that's near and dear to my heart, so I had to write about it.https://t.co/sT0FU4xPCb
“Because open-source is good at cutting through the bullshit, it also makes it an equalizer. If you come from an underprivileged background, you should absolutely use GitHub to get ahead. That's exactly what I did. I owe my entire career to open-source.” https://t.co/RggNx46PF5
— hardmaru (@hardmaru) June 10, 2018
Thread on Scientific Research
(Long thread. Click on the tweet to read the full thread.)
Whenever I criticize professors I get email from fellow profs admonishing me. Today it was the charge that I was "...attacking the people who have power, rather than diplomatically trying to change their minds".
— Chris Chambers (@chrisdc77) June 10, 2018
Dear Profs: This thread is my autoreply & my only reply. /1 https://t.co/NyD4pCdyzs