Improving Covariance Conditioning of the SVD Meta-layer by Orthogonality

• URL: http://arxiv.org/abs/2207.02119v1
• Date: Tue, 5 Jul 2022 15:39:29 GMT
• Title: Improving Covariance Conditioning of the SVD Meta-layer by Orthogonality
• Authors: Yue Song, Nicu Sebe, Wei Wang
• Abstract summary: Nearest Orthogonal Gradient (NOG) and Optimal Learning Rate (OLR) are proposed. Experiments on visual recognition demonstrate that our methods can simultaneously improve the covariance conditioning and generalization.
• Score: 65.67315418971688
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Inserting an SVD meta-layer into neural networks is prone to make the covariance ill-conditioned, which could harm the model in the training stability and generalization abilities. In this paper, we systematically study how to improve the covariance conditioning by enforcing orthogonality to the Pre-SVD layer. Existing orthogonal treatments on the weights are first investigated. However, these techniques can improve the conditioning but would hurt the performance. To avoid such a side effect, we propose the Nearest Orthogonal Gradient (NOG) and Optimal Learning Rate (OLR). The effectiveness of our methods is validated in two applications: decorrelated Batch Normalization (BN) and Global Covariance Pooling (GCP). Extensive experiments on visual recognition demonstrate that our methods can simultaneously improve the covariance conditioning and generalization. Moreover, the combinations with orthogonal weight can further boost the performances.

Related papers

Err

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.