Towards Effective Data Augmentations via Unbiased GAN Utilization
- Publication Type:
- Conference Proceeding
- Citation:
- Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2019, 11672 LNAI pp. 555 - 567
- Issue Date:
- 2019-01-01
Closed Access
| Filename | Description | Size | |||
|---|---|---|---|---|---|
| 2019_Book_PRICAI2019TrendsInArtificialIn1.pdf | Published version | 271.84 kB |
Copyright Clearance Process
- Recently Added
- In Progress
- Closed Access
This item is closed access and not available.
© 2019, Springer Nature Switzerland AG. The parameters of any machine learning (ML) model are obtained from the dataset on which the model is trained. However, existing research reveals that many datasets appear to have strong build-in biases. These biases are inherently learned by the learning mechanism of the ML model which adversely affects their generalization performance. In this research, we propose a new supervised data augmentation mechanism which we call as Data Augmentation Pursuit (DAP). The DAP generates labelled synthetic data instances for augmenting the raw datasets. To demonstrate the effectiveness of utilizing DAP for reducing model bias, we perform comprehensive experiments on real world image dataset. CNN models trained on augmented dataset obtained using DAP achieves significantly better classification performance and exhibits reduction in the bias learned by their learning mechanism.
Please use this identifier to cite or link to this item: