An Overview of Prototype Formulations for Interpretable Deep Learning
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_ | 1866914238332665856 |
|---|---|
| author | Li, Maximilian Xiling Rudolf, Korbinian Franz Mattes, Paul Blank, Nils Lioutikov, Rudolf |
| author_facet | Li, Maximilian Xiling Rudolf, Korbinian Franz Mattes, Paul Blank, Nils Lioutikov, Rudolf |
| contents | Prototypical part networks offer interpretable alternatives to black-box deep learning models by learning visual prototypes for classification. This work provides a comprehensive analysis of prototype formulations, comparing point-based and probabilistic approaches in both Euclidean and hyperspherical latent spaces.
We introduce HyperPG, a probabilistic prototype representation using Gaussian distributions on hyperspheres. Experiments on CUB-200-2011, Stanford Cars, and Oxford Flowers datasets show that hyperspherical prototypes outperform standard Euclidean formulations. Critically, hyperspherical prototypes maintain competitive performance under simplified training schemes, while Euclidean prototypes require extensive hyperparameter tuning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_08925 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | An Overview of Prototype Formulations for Interpretable Deep Learning Li, Maximilian Xiling Rudolf, Korbinian Franz Mattes, Paul Blank, Nils Lioutikov, Rudolf Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition Prototypical part networks offer interpretable alternatives to black-box deep learning models by learning visual prototypes for classification. This work provides a comprehensive analysis of prototype formulations, comparing point-based and probabilistic approaches in both Euclidean and hyperspherical latent spaces. We introduce HyperPG, a probabilistic prototype representation using Gaussian distributions on hyperspheres. Experiments on CUB-200-2011, Stanford Cars, and Oxford Flowers datasets show that hyperspherical prototypes outperform standard Euclidean formulations. Critically, hyperspherical prototypes maintain competitive performance under simplified training schemes, while Euclidean prototypes require extensive hyperparameter tuning. |
| title | An Overview of Prototype Formulations for Interpretable Deep Learning |
| topic | Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2410.08925 |