Reasons like these are why I get frustrated when people claim the only reason why NNs are popular now is because of computational advances. This isn't true - th

Reasons like these are why I get frustrated when people claim the only reason why NNs are popular now is because of computational advances. This isn't true - there is lots of little learnings that have been developed in recent time to make training these networks reliable in production. Large companies (and universities) have had access to high performance computers and clusters to train interesting networks (say char-rnn) for some time.