When someone says "we need theory of deep learning", note that probably nothing will "count" as a theory of deep learning unless it is *their* theory, or unless it is speaking a language, and using techniques, that they already have a bias towards.
NYU's Andrew Gordon Wilson argues that demands for a unifying deep learning theory are biased by researchers' preferred frameworks
Researchers rarely accept theories outside their preferred mathematical languages.
Most Activity
When someone says "we need theory of deep learning", note that probably nothing will "count" as a theory of deep learning unless it is *their* theory, or unless it is speaking a language, and using techniques, that they already have a bias towards.