Deep neural network structures solving variational inequalities (Q2194605): Difference between revisions
From MaRDI portal
Created a new Item |
Added link to MaRDI item. |
||
| links / mardi / name | links / mardi / name | ||
Revision as of 01:36, 2 February 2024
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Deep neural network structures solving variational inequalities |
scientific article |
Statements
Deep neural network structures solving variational inequalities (English)
0 references
4 September 2020
0 references
The authors study nonlinear composite models alternating proximity and affine operators showing in particular that a wide range of activation operators used in neural networks are in fact proximity operators. Some asymptotic properties of the resulting process are provided.
0 references
averaged operator
0 references
deep neural network
0 references
monotone operator
0 references
nonexpansive operator
0 references
proximity operator
0 references
variational inequality
0 references
