AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Do You Think It's a Good Thing to Have?
If you train a discreet algrithm on top of this continuous function, it will be able to learn in the same continuous way that. a neurow network learns - but can self modify its discreet algarithmand. I think lacoon would rather have something which is entirely differential and learnable as part of the main model. But i don't necessarily, e he's talking about an all we need, like we need to end to end learn like a machine where you simply in, put something and then out pops through forward, prop outo hops and end.