Reinforcing Pre-trained Models Using Counterfactual Images
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866929391903178752 |
|---|---|
| author | Li, Xiang Togo, Ren Maeda, Keisuke Ogawa, Takahiro Haseyama, Miki |
| author_facet | Li, Xiang Togo, Ren Maeda, Keisuke Ogawa, Takahiro Haseyama, Miki |
| contents | This paper proposes a novel framework to reinforce classification models using language-guided generated counterfactual images. Deep learning classification models are often trained using datasets that mirror real-world scenarios. In this training process, because learning is based solely on correlations with labels, there is a risk that models may learn spurious relationships, such as an overreliance on features not central to the subject, like background elements in images. However, due to the black-box nature of the decision-making process in deep learning models, identifying and addressing these vulnerabilities has been particularly challenging. We introduce a novel framework for reinforcing the classification models, which consists of a two-stage process. First, we identify model weaknesses by testing the model using the counterfactual image dataset, which is generated by perturbed image captions. Subsequently, we employ the counterfactual images as an augmented dataset to fine-tune and reinforce the classification model. Through extensive experiments on several classification models across various datasets, we revealed that fine-tuning with a small set of counterfactual images effectively strengthens the model. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_13316 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Reinforcing Pre-trained Models Using Counterfactual Images Li, Xiang Togo, Ren Maeda, Keisuke Ogawa, Takahiro Haseyama, Miki Computer Vision and Pattern Recognition Multimedia This paper proposes a novel framework to reinforce classification models using language-guided generated counterfactual images. Deep learning classification models are often trained using datasets that mirror real-world scenarios. In this training process, because learning is based solely on correlations with labels, there is a risk that models may learn spurious relationships, such as an overreliance on features not central to the subject, like background elements in images. However, due to the black-box nature of the decision-making process in deep learning models, identifying and addressing these vulnerabilities has been particularly challenging. We introduce a novel framework for reinforcing the classification models, which consists of a two-stage process. First, we identify model weaknesses by testing the model using the counterfactual image dataset, which is generated by perturbed image captions. Subsequently, we employ the counterfactual images as an augmented dataset to fine-tune and reinforce the classification model. Through extensive experiments on several classification models across various datasets, we revealed that fine-tuning with a small set of counterfactual images effectively strengthens the model. |
| title | Reinforcing Pre-trained Models Using Counterfactual Images |
| topic | Computer Vision and Pattern Recognition Multimedia |
| url | https://arxiv.org/abs/2406.13316 |