Data Pruning via Separability, Integrity, and Model Uncertainty-Aware Importance Sampling
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916405555757056 |
|---|---|
| author | Grosz, Steven Zhao, Rui Ranjan, Rajeev Wang, Hongcheng Aggarwal, Manoj Medioni, Gerard Jain, Anil |
| author_facet | Grosz, Steven Zhao, Rui Ranjan, Rajeev Wang, Hongcheng Aggarwal, Manoj Medioni, Gerard Jain, Anil |
| contents | This paper improves upon existing data pruning methods for image classification by introducing a novel pruning metric and pruning procedure based on importance sampling. The proposed pruning metric explicitly accounts for data separability, data integrity, and model uncertainty, while the sampling procedure is adaptive to the pruning ratio and considers both intra-class and inter-class separation to further enhance the effectiveness of pruning. Furthermore, the sampling method can readily be applied to other pruning metrics to improve their performance. Overall, the proposed approach scales well to high pruning ratio and generalizes better across different classification models, as demonstrated by experiments on four benchmark datasets, including the fine-grained classification scenario. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_13915 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Data Pruning via Separability, Integrity, and Model Uncertainty-Aware Importance Sampling Grosz, Steven Zhao, Rui Ranjan, Rajeev Wang, Hongcheng Aggarwal, Manoj Medioni, Gerard Jain, Anil Computer Vision and Pattern Recognition This paper improves upon existing data pruning methods for image classification by introducing a novel pruning metric and pruning procedure based on importance sampling. The proposed pruning metric explicitly accounts for data separability, data integrity, and model uncertainty, while the sampling procedure is adaptive to the pruning ratio and considers both intra-class and inter-class separation to further enhance the effectiveness of pruning. Furthermore, the sampling method can readily be applied to other pruning metrics to improve their performance. Overall, the proposed approach scales well to high pruning ratio and generalizes better across different classification models, as demonstrated by experiments on four benchmark datasets, including the fine-grained classification scenario. |
| title | Data Pruning via Separability, Integrity, and Model Uncertainty-Aware Importance Sampling |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2409.13915 |