The Gradient: Perspectives on AI cover image

Kyunghyun Cho: Neural Machine Translation, Language, and Doing Good Science

The Gradient: Perspectives on AI

00:00

Is There a Way to Predict the Model's Output?

There are people in the reinforcement learning and open-endedness communities who are exploring some of those ideas. I also thought what you said about your counterfactual analysis was interesting just because looking at some other work we've seen recently. So these are all quite all the attempts and hopefully an idea over the next few years or so we'll have much, let's say, well-founded, well-grounded and well-principled platforms for advanced evaluation.

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app