You can feed books to an RNN and it learn how to produce text.

What I'm interested in is an algorithm that, given say 20 letters it suggest, say the best 10 options for the next 10 letters.

So for example it begins with "The cat jumped " and then we get various options such as "over the dog". "on the table" and so on.

My initial thoughts are to first use the most likely next letters. Then find the letter which is most uncertain and change this to the second likely next letter. And repeat this process.

(Then I may have another evaluation neural network to assess which is "best" English.)

In other words I want the RNN to "think ahead" at what it's saying - much like a chess playing machine.

| improve this question | | | | |
  • $\begingroup$ I'd recommend going through the CBOW-based word2vec, which pretty much does the same thing: israelg99.github.io/2017-03-23-Word2Vec-Explained $\endgroup$ – Syed Ali Hamza Aug 17 '19 at 17:17
  • $\begingroup$ @Syed Thanks I'll take a look $\endgroup$ – zooby Aug 17 '19 at 18:25
  • $\begingroup$ you mention neural text generations which work off of sampling (so just sample 10x) or find 10 most likely $\endgroup$ – mshlis Aug 19 '19 at 2:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.