Training neural networks for and by interpolation
In modern supervised learning, many deep neural networks are able to interpolate the data: the empirical loss can be driven to near zero on all samples simultaneously. In this work, we explicitly exploit this interpolation property for the design of a new optimization algorithm for deep learning, wh...
Váldodahkkit: | Berrada, L, Zisserman, A, Kumar, MP |
---|---|
Materiálatiipa: | Conference item |
Giella: | English |
Almmustuhtton: |
Journal of Machine Learning Research
2020
|
Geahča maid
-
Deep Frank-Wolfe for neural network optimization
Dahkki: Berrada, L, et al.
Almmustuhtton: (2018) -
Deep Frank-Wolfe for neural network optimization
Dahkki: Berrada, L, et al.
Almmustuhtton: (2019) -
Smooth loss functions for deep top-k classification
Dahkki: Berrada, L, et al.
Almmustuhtton: (2018) -
Smooth loss functions for deep top-k classification
Dahkki: Berrada, L, et al.
Almmustuhtton: (2018) -
Trusting SVM for piecewise linear CNNs
Dahkki: Berrada, L, et al.
Almmustuhtton: (2016)