Smooth Kolmogorov Arnold networks enabling structural knowledge representation
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_ | 1866929358968455168 |
|---|---|
| author | Samadi, Moein E. Müller, Younes Schuppert, Andreas |
| author_facet | Samadi, Moein E. Müller, Younes Schuppert, Andreas |
| contents | Kolmogorov-Arnold Networks (KANs) offer an efficient and interpretable alternative to traditional multi-layer perceptron (MLP) architectures due to their finite network topology. However, according to the results of Kolmogorov and Vitushkin, the representation of generic smooth functions by KAN implementations using analytic functions constrained to a finite number of cutoff points cannot be exact. Hence, the convergence of KAN throughout the training process may be limited. This paper explores the relevance of smoothness in KANs, proposing that smooth, structurally informed KANs can achieve equivalence to MLPs in specific function classes. By leveraging inherent structural knowledge, KANs may reduce the data required for training and mitigate the risk of generating hallucinated predictions, thereby enhancing model reliability and performance in computational biomedicine. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_11318 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Smooth Kolmogorov Arnold networks enabling structural knowledge representation Samadi, Moein E. Müller, Younes Schuppert, Andreas Machine Learning Disordered Systems and Neural Networks Artificial Intelligence Kolmogorov-Arnold Networks (KANs) offer an efficient and interpretable alternative to traditional multi-layer perceptron (MLP) architectures due to their finite network topology. However, according to the results of Kolmogorov and Vitushkin, the representation of generic smooth functions by KAN implementations using analytic functions constrained to a finite number of cutoff points cannot be exact. Hence, the convergence of KAN throughout the training process may be limited. This paper explores the relevance of smoothness in KANs, proposing that smooth, structurally informed KANs can achieve equivalence to MLPs in specific function classes. By leveraging inherent structural knowledge, KANs may reduce the data required for training and mitigate the risk of generating hallucinated predictions, thereby enhancing model reliability and performance in computational biomedicine. |
| title | Smooth Kolmogorov Arnold networks enabling structural knowledge representation |
| topic | Machine Learning Disordered Systems and Neural Networks Artificial Intelligence |
| url | https://arxiv.org/abs/2405.11318 |