Title | Complexity Control by Gradient Descent in Deep Networks |
Publication Type | Journal Article |
Year of Publication | 2020 |
Authors | Poggio, T, Liao, Q, Banburski, A |
Journal | Nature Communications |
Volume | 11 |
Date Published | 02/2020 |
Abstract | Overparametrized deep network predict well despite the lack of an explicit complexity control during training such as an explicit regularization term. For exponential-type loss functions, we solve this puzzle by showing an effective regularization effect of gradient descent in terms of the normalized weights that are relevant for classification. |
URL | https://www.nature.com/articles/s41467-020-14663-9 |
DOI | 10.1038/s41467-020-14663-9 |
Download:
s41467-020-14663-9.pdf
Associated Module:
CBMM Relationship:
- CBMM Funded