Interpretable global minima of deep ReLU neural networks on sequentially separable data
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917196781846528 |
|---|---|
| author | Chen, Thomas Ewald, Patrícia Muñoz |
| author_facet | Chen, Thomas Ewald, Patrícia Muñoz |
| contents | We explicitly construct zero loss neural network classifiers. We write the weight matrices and bias vectors in terms of cumulative parameters, which determine truncation maps acting recursively on input space. The configurations for the training data considered are (i) sufficiently small, well separated clusters corresponding to each class, and (ii) equivalence classes which are sequentially linearly separable. In the best case, for $Q$ classes of data in $\mathbb{R}^M$, global minimizers can be described with $Q(M+2)$ parameters. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_07098 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Interpretable global minima of deep ReLU neural networks on sequentially separable data Chen, Thomas Ewald, Patrícia Muñoz Machine Learning Artificial Intelligence Mathematical Physics Optimization and Control 57R70, 62M45 We explicitly construct zero loss neural network classifiers. We write the weight matrices and bias vectors in terms of cumulative parameters, which determine truncation maps acting recursively on input space. The configurations for the training data considered are (i) sufficiently small, well separated clusters corresponding to each class, and (ii) equivalence classes which are sequentially linearly separable. In the best case, for $Q$ classes of data in $\mathbb{R}^M$, global minimizers can be described with $Q(M+2)$ parameters. |
| title | Interpretable global minima of deep ReLU neural networks on sequentially separable data |
| topic | Machine Learning Artificial Intelligence Mathematical Physics Optimization and Control 57R70, 62M45 |
| url | https://arxiv.org/abs/2405.07098 |