Training an AI to write BuzzFeed clickbait headlines (opens in new tab)

(dataskunkworks.com)

2 pointsspiggytopes7y ago1 comments

1 comments

1 comments · 1 top-level

MR4D7y ago

Am I missing something here, or would a mark of chain be much easier to write and solve the problem just as well?

Seems like much less effort. Of course, maybe the author wanted to work on an RNN instead?

j / k navigate · click thread line to collapse