Wednesday, April 26, 2017

Review for 26 April 2017

Below are some of the interesting links I Tweeted about recently.

  1. Couples don't need ML to tell them when they are about to fight they need help to reduce stresses that cause fights:
  2. Facebook open sources it's preferred deep learning framework, Caffe2: 
  3. Five important machine learning projects: 
  4. Cheap learning, not deep learning-simple algorithms are still useful for many problems: 
  5. Neural networks predict heart attacks better than doctors: 
  6. "Automation is a great excuse for assholery", or, why biased data produces biased models: 
  7. MS has merged Python into SQL Server 2017 to give users access to machine learning frameworks on the server: 
  8. Why using AI in the court system is a bad idea:
  9. Thoughts by neuroscientists on Elon Musk's Neural Lace:
  10. 95% of engineers graduating Indian institutions are unfit for software development jobs:
  11. Soft Computing, Volume 21, Number 9, May 2017 
  12. How Facebook plans to make more money using machine learning: 
  13. AI investment is up-we are definitely in an AI hype-cycle: 
  14. Introduction to transfer learning: 
  15. Machine learning moves into recruitment:  Like there's no bias in *those* models..