Jonathan Frankle, MIT: The lottery ticket hypothesis and the science of deep learning
Generally Intelligent
00:00
Pruning Networks of Initialization: A Review
Next is concerned that the literature is claiming victory in an unjustified way./nNext is trying to develop techniques for pruning networks of initialization, but is finding that the results are not always good./nNext is concerned that the literature is not doing good science, and is cherry picking settings or sparsities to claim victory where they can.
Play episode from 27:40
Transcript
