Weight Rotation as a Regularization Strategy in Convolutional Neural Networks

• Type: Article in International Conference Proceedings Book
• Date: 2019

Authors

Eduardo Castro (Author) Other; José Costa Pereira (Author) Other; Jaime S. Cardoso (Author) FEUP

Conference proceedings International

Title: Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS

Pages: 2106-2110

41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2019

23 July 2019 through 27 July 2019

Indexing

ISI Web of Knowledge - 0 Citations

Scopus - 0 Citations

Other information

Authenticus ID: P-00R-JD7

DOI: 10.1109/embc.2019.8856448

Abstract (EN): Convolutional Neural Networks (CNN) have become the gold standard in many visual recognition tasks including medical applications. Due to their high variance, however, these models are prone to over-fit the data they are trained on. To mitigate this problem, one of the most common strategies, is to perform data augmentation. Rotation, scaling and translation are common operations. In this work we propose an alternative method to rotation-based data augmentation where the rotation transformation is performed inside the CNN architecture. In each training batch the weights of all convolutional layers are rotated by the same random angle. We validate our proposed method empirically showing its usefulness under different scenarios.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 5