Data structure > labels? Unsupervised heuristics for SVM hyperparameter estimation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2021
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866910339560374272 |
|---|---|
| author | Cholewa, Michał Romaszewski, Michał Głomb, Przemysław |
| author_facet | Cholewa, Michał Romaszewski, Michał Głomb, Przemysław |
| contents | Classification is one of the main areas of pattern recognition research, and within it, Support Vector Machine (SVM) is one of the most popular methods outside of field of deep learning -- and a de-facto reference for many Machine Learning approaches. Its performance is determined by parameter selection, which is usually achieved by a time-consuming grid search cross-validation procedure (GSCV). That method, however relies on the availability and quality of labelled examples and thus, when those are limited can be hindered. To address that problem, there exist several unsupervised heuristics that take advantage of the characteristics of the dataset for selecting parameters instead of using class label information. While an order of magnitude faster, they are scarcely used under the assumption that their results are significantly worse than those of grid search. To challenge that assumption, we have proposed improved heuristics for SVM parameter selection and tested it against GSCV and state of the art heuristics on over 30 standard classification datasets. The results show not only its advantage over state-of-art heuristics but also that it is statistically no worse than GSCV. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2111_02164 |
| institution | arXiv |
| publishDate | 2021 |
| record_format | arxiv |
| spellingShingle | Data structure > labels? Unsupervised heuristics for SVM hyperparameter estimation Cholewa, Michał Romaszewski, Michał Głomb, Przemysław Machine Learning 68T05 I.2.6 Classification is one of the main areas of pattern recognition research, and within it, Support Vector Machine (SVM) is one of the most popular methods outside of field of deep learning -- and a de-facto reference for many Machine Learning approaches. Its performance is determined by parameter selection, which is usually achieved by a time-consuming grid search cross-validation procedure (GSCV). That method, however relies on the availability and quality of labelled examples and thus, when those are limited can be hindered. To address that problem, there exist several unsupervised heuristics that take advantage of the characteristics of the dataset for selecting parameters instead of using class label information. While an order of magnitude faster, they are scarcely used under the assumption that their results are significantly worse than those of grid search. To challenge that assumption, we have proposed improved heuristics for SVM parameter selection and tested it against GSCV and state of the art heuristics on over 30 standard classification datasets. The results show not only its advantage over state-of-art heuristics but also that it is statistically no worse than GSCV. |
| title | Data structure > labels? Unsupervised heuristics for SVM hyperparameter estimation |
| topic | Machine Learning 68T05 I.2.6 |
| url | https://arxiv.org/abs/2111.02164 |