The Gradient: Perspectives on AI cover image

Jeremie Harris: Realistic Alignment and AI Policy

The Gradient: Perspectives on AI

CHAPTER

The Importance of Inference in Deep Neural Networks

When we try to do theory work on deep neural networks, a lot of classical machine learning theory tends to break apart. I think there are important differences here so I don't know how much is inference holds up but it's notable when we think about the idea that certain people would like something that is provable justifiable. When it comes to some alignment results, we might want when I'm just analyzing the behavior of a neural network on its own, not considering alignment. That's something we still don't have a handle on.

00:00
Transcript
Play full episode

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
App store bannerPlay store banner