115 citations · 1.1k across the 38 of their papers we have counts for
Showing cs.LGShow all
2 papers · 1 filter
cs.LG2020★ 5 cited
Partial local entropy and anisotropy in deep weight spaces
Daniele Musso
We refine a recently-proposed class of local entropic loss functions by restricting the smoothening regularization to only a subset of weights. The new loss functions are referred…
cs.LG2020★ 3 cited
Stochastic gradient descent with random learning rate
Daniele Musso
We propose to optimize neural networks with a uniformly-distributed random learning rate. The associated stochastic gradient descent algorithm can be approximated by continuous sto…