AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Is There Any Such Thing as General Robustness?
There's no such thing as general robustness, because i got composed two different assumptions about the world were in one of em. So you have a set of people that are just kind of doing the deep learning thing, which diction lets you get away with it. In a deep linferis not actually addressing causality in any kind of sound way and fool a reviewer, but not sarily be doing it. And so i tend to think that, like a lot o's other problems are kind of more foundational. They're not problems where like, we know how to evaluate systems,. loks that people try stuff and whatever their problems were.