CIFAR-10-Warehouse: Broad and More Realistic Testbeds in Model Generalization Analysis
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2023
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916158112792576 |
|---|---|
| author | Sun, Xiaoxiao Leng, Xingjian Wang, Zijian Yang, Yang Huang, Zi Zheng, Liang |
| author_facet | Sun, Xiaoxiao Leng, Xingjian Wang, Zijian Yang, Yang Huang, Zi Zheng, Liang |
| contents | Analyzing model performance in various unseen environments is a critical research problem in the machine learning community. To study this problem, it is important to construct a testbed with out-of-distribution test sets that have broad coverage of environmental discrepancies. However, existing testbeds typically either have a small number of domains or are synthesized by image corruptions, hindering algorithm design that demonstrates real-world effectiveness. In this paper, we introduce CIFAR-10-Warehouse, consisting of 180 datasets collected by prompting image search engines and diffusion models in various ways. Generally sized between 300 and 8,000 images, the datasets contain natural images, cartoons, certain colors, or objects that do not naturally appear. With CIFAR-10-W, we aim to enhance the evaluation and deepen the understanding of two generalization tasks: domain generalization and model accuracy prediction in various out-of-distribution environments. We conduct extensive benchmarking and comparison experiments and show that CIFAR-10-W offers new and interesting insights inherent to these tasks. We also discuss other fields that would benefit from CIFAR-10-W. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2310_04414 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | CIFAR-10-Warehouse: Broad and More Realistic Testbeds in Model Generalization Analysis Sun, Xiaoxiao Leng, Xingjian Wang, Zijian Yang, Yang Huang, Zi Zheng, Liang Computer Vision and Pattern Recognition Analyzing model performance in various unseen environments is a critical research problem in the machine learning community. To study this problem, it is important to construct a testbed with out-of-distribution test sets that have broad coverage of environmental discrepancies. However, existing testbeds typically either have a small number of domains or are synthesized by image corruptions, hindering algorithm design that demonstrates real-world effectiveness. In this paper, we introduce CIFAR-10-Warehouse, consisting of 180 datasets collected by prompting image search engines and diffusion models in various ways. Generally sized between 300 and 8,000 images, the datasets contain natural images, cartoons, certain colors, or objects that do not naturally appear. With CIFAR-10-W, we aim to enhance the evaluation and deepen the understanding of two generalization tasks: domain generalization and model accuracy prediction in various out-of-distribution environments. We conduct extensive benchmarking and comparison experiments and show that CIFAR-10-W offers new and interesting insights inherent to these tasks. We also discuss other fields that would benefit from CIFAR-10-W. |
| title | CIFAR-10-Warehouse: Broad and More Realistic Testbeds in Model Generalization Analysis |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2310.04414 |