GPT-3 + GAN

post by stick109 · 2020-10-17T07:58:42.839Z · LW · GW · 1 comment

This is a question post.

Contents

  Answers
    1 Alexander Mathiasen
    1 Aiyen
None
1 comment

I am wondering if anyone tried to combine GPT-3 with GAN, basically trying to train network that would feed GPT-3 questions and then judge responses as correct/incorrect, thus providing GPT-3 opportunity to improve.

Does my question even make sense or I am far off base?

Answers

answer by Alexander Mathiasen · 2021-09-22T14:34:10.533Z · LW(p) · GW(p)

This would require you to sample from GPT during training. If you want a sentence with 500 words you need to evaluate GPT 500 times. As a result, it would slow down training 500 times. The clever thing with GPT (and other autoregressive models) is that they circumvent sampling during training!

answer by Aiyen · 2020-10-19T23:00:49.135Z · LW(p) · GW(p)

That might work.  Maybe have the adversarial network try to distinguish GPT-3 text from human text?  That said, GPT-3 is already trying to predict humanlike text continuations, so there's a decent chance that having a separate GAN layer wouldn't help.  It's probably worth doing the experiment though; traditional GANs work by improving the discriminator as well as the desired categorizer, so there's a chance it could work here too. 

1 comment

Comments sorted by top scores.