The Problem of Basical Attribution in Machine Learning Networks

When you have machine learning networks that basically take all these factors and plug them into each other and map them to some outcome, it's very difficult to trace back through the network. We don't know exactly what we want to do at the individual a, ameno a base per level, ameno asid base. So so, a, we could make guesses, but, ah, you could make mistakes, and it's because we don't knowExactly what's causal and what's notand with these things, how do you bound the kind of certaint involved in making these edits?

Transcript

Play full episode

Remember Everything You Learn from Podcasts

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