2 papers
cs.LG2020
Unique Properties of Flat Minima in Deep Networks
Rotem Mulayoff, Tomer Michaeli
It is well known that (stochastic) gradient descent has an implicit bias towards flat minima. In deep neural network training, this mechanism serves to screen out minima. However,…
eess.SP2018
On the Minimal Overcompleteness Allowing Universal Sparse Representation
Rotem Mulayoff, Tomer Michaeli
Sparse representation over redundant dictionaries constitutes a good model for many classes of signals (e.g., patches of natural images, segments of speech signals, etc.). However,…