Sign in

Unveiling the structure of wide flat minima in neural networks

By Carlo Baldassi and others
The success of deep learning has revealed the application potential of neural networks across the sciences and opened up fundamental theoretical problems. In particular, the fact that learning algorithms based on simple variants of gradient methods are able to find near-optimal minima of highly nonconvex loss functions is an unexpected... Show more
October 26, 2021
=
0
Loading PDF…
Loading full text...
Similar articles
Loading recommendations...
=
0
x1
Unveiling the structure of wide flat minima in neural networks
Click on play to start listening