AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
How to Increase Your Observability in Neural Networks?
The ways that were able to represent graphs in neural networks, they're overly constrained from the richness that you might want to express them as a mathematician. In particular, i'm even willing to go one level higher and say, if you think at large, at machine learning people, and i'm one of them, we don't really question where the data came from. It's just, here's data. Here's use a prediction task. And perhaps for five minutes or ten minutes, your algualem is the best predictive Algualem on planet,. until the next person that comes in with their own method.