News for hackers. Stuff that matters.
Re:Ask HN: If deep learning is so effective why dont Boston ... by opless on Monday August 17, 2015 @08:01PM (#10076877)
Deep learning is just a fad.

Throw enough data at your favourite ML algorithm and you'll get better results.

Re:Ask HN: If deep learning is so effective why dont Boston ... by skythomas on Monday August 17, 2015 @08:33PM (#10076986)
Uhm.. no. What is special today about deep learning is the hierarchy of features. Yes, we are talking primarily about CNN's and RNN's trained with SGD TODAY. Yes,the entire catalog of ML techniques are still very interesting. Yes, in the future many of these techniques will be viable in a deep setting. But, no, you don't just get state of the art results by using linear regression or HOG and SIFT SVM's on a big data set.

Hierarchy of features is important.

Re:Ask HN: If deep learning is so effective why dont Boston ... by wodenokoto on Monday August 17, 2015 @10:48PM (#10077348)
No, that's actually the main benefit of NN. They improve with more data for longer than almost any alternative algorithm.