Hacker News new | ask | show | jobs
by jl2718 2848 days ago
Statistically-random excellent performance in complex tasks is very unlikely. More likely is that examples are in-sample from a small training set or very similar. Big NNs can memorize anything.
1 comments

They can memorize any supervised learning task, but so far, we haven't been able to see any deep generative model successfully memorize something more complex than MNIST.