Test-Time Mixup Augmentation for Data and Class-Specific Uncertainty Estimation in Deep Learning Image Classification
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2022
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866929242875363328 |
|---|---|
| author | Lee, Hansang Lee, Haeil Hong, Helen Kim, Junmo |
| author_facet | Lee, Hansang Lee, Haeil Hong, Helen Kim, Junmo |
| contents | Uncertainty estimation of trained deep learning networks is valuable for optimizing learning efficiency and evaluating the reliability of network predictions. In this paper, we propose a method for estimating uncertainty in deep learning image classification using test-time mixup augmentation (TTMA). To improve the ability to distinguish correct and incorrect predictions in existing aleatoric uncertainty, we introduce TTMA data uncertainty (TTMA-DU) by applying mixup augmentation to test data and measuring the entropy of the predicted label histogram. In addition to TTMA-DU, we propose TTMA class-specific uncertainty (TTMA-CSU), which captures aleatoric uncertainty specific to individual classes and provides insight into class confusion and class similarity within the trained network. We validate our proposed methods on the ISIC-18 skin lesion diagnosis dataset and the CIFAR-100 real-world image classification dataset. Our experiments show that (1) TTMA-DU more effectively differentiates correct and incorrect predictions compared to existing uncertainty measures due to mixup perturbation, and (2) TTMA-CSU provides information on class confusion and class similarity for both datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2212_00214 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Test-Time Mixup Augmentation for Data and Class-Specific Uncertainty Estimation in Deep Learning Image Classification Lee, Hansang Lee, Haeil Hong, Helen Kim, Junmo Computer Vision and Pattern Recognition Artificial Intelligence Uncertainty estimation of trained deep learning networks is valuable for optimizing learning efficiency and evaluating the reliability of network predictions. In this paper, we propose a method for estimating uncertainty in deep learning image classification using test-time mixup augmentation (TTMA). To improve the ability to distinguish correct and incorrect predictions in existing aleatoric uncertainty, we introduce TTMA data uncertainty (TTMA-DU) by applying mixup augmentation to test data and measuring the entropy of the predicted label histogram. In addition to TTMA-DU, we propose TTMA class-specific uncertainty (TTMA-CSU), which captures aleatoric uncertainty specific to individual classes and provides insight into class confusion and class similarity within the trained network. We validate our proposed methods on the ISIC-18 skin lesion diagnosis dataset and the CIFAR-100 real-world image classification dataset. Our experiments show that (1) TTMA-DU more effectively differentiates correct and incorrect predictions compared to existing uncertainty measures due to mixup perturbation, and (2) TTMA-CSU provides information on class confusion and class similarity for both datasets. |
| title | Test-Time Mixup Augmentation for Data and Class-Specific Uncertainty Estimation in Deep Learning Image Classification |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2212.00214 |