End-to-end training of deep kernel map networks for image classification

by   Mingyuan Jiu, et al.

Deep kernel map networks have shown excellent performances in various classification problems including image annotation. Their general recipe consists in aggregating several layers of singular value decompositions (SVDs) – that map data from input spaces into high dimensional spaces – while preserving the similarity of the underlying kernels. However, the potential of these deep map networks has not been fully explored as the original setting of these networks focuses mainly on the approximation quality of their kernels and ignores their discrimination power. In this paper, we introduce a novel "end-to-end" design for deep kernel map learning that balances the approximation quality of kernels and their discrimination power. Our method proceeds in two steps; first, layerwise SVD is applied in order to build initial deep kernel map approximations and then an "end-to-end" supervised learning is employed to further enhance their discrimination power while maintaining their efficiency. Extensive experiments, conducted on the challenging ImageCLEF annotation benchmark, show the high efficiency and the out-performance of this two-step process with respect to different related methods.


Learning Deep Context-Network Architectures for Image Annotation

Context plays an important role in visual pattern recognition as it prov...

Deep Context-Aware Kernel Networks

Context plays a crucial role in visual recognition as it provides comple...

Learning Kernel for Conditional Moment-Matching Discrepancy-based Image Classification

Conditional Maximum Mean Discrepancy (CMMD) can capture the discrepancy ...

Learning Explicit Deep Representations from Deep Kernel Networks

Deep kernel learning aims at designing nonlinear combinations of multipl...

Exploring Kernel Functions in the Softmax Layer for Contextual Word Classification

Prominently used in support vector machines and logistic regressions, ke...

End-to-End Supervised Multilabel Contrastive Learning

Multilabel representation learning is recognized as a challenging proble...

Generative Autoencoder Kernels on Deep Learning for Brain Activity Analysis

Deep Learning (DL) is a two-step classification model that consists feat...

Please sign up or login with your details

Forgot password? Click here to reset