Hypothesis (& one I’m pretty sure is true): adversarial examples will be a permanent problem in deep learning/ML. They exist because these systems do not have an underlying model that corresponds to their behavior.https://twitter.com/catherineols/status/1020458649825636352 …
I don’t think that the amount of training data needs to be much larger, but the data has to be largely mappable to the same set of relations, ie a single cohesive world simulation. I suppose we agree on that?
-
-
I don't know what you mean by "single cohesive world simulation".
-
The world that we subjectively inhabit is a simulation generated by the neocortex, which is tuned to predict the sparse noisy samples we receive from our sensory neurons.
- 3 more replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.