Can Machine Learning Replace Signal Processing?

Prof. Nathan Intrator
Israel, Blavatnik School of Computer Science, Sagol School of Neuroscience
Recently, deep learning has greatly improved error rates, which were stagnant over the last decade.
A question arises about whether learning itself can obtain optimal data representation in problems traditionally thought to be in the signal processing realm.
The talk will describe relevant examples from different domains and viewpoints, and address the question from a brain research perspective.