Timezone: »

 
Spotlight
Connecting Sphere Manifolds Hierarchically for Regularization
Damien Scieur · Youngsung Kim

Tue Jul 20 06:30 AM -- 06:35 AM (PDT) @
in Deep Learning Algorithms 2 »

This paper considers classification problems with hierarchically organized classes. We force the classifier (hyperplane) of each class to belong to a sphere manifold, whose center is the classifier of its super-class. Then, individual sphere manifolds are connected based on their hierarchical relations. Our technique replaces the last layer of a neural network by combining a spherical fully-connected layer with a hierarchical layer. This regularization is shown to improve the performance of widely used deep neural network architectures (ResNet and DenseNet) on publicly available datasets (CIFAR100, CUB200, Stanford dogs, Stanford cars, and Tiny-ImageNet).

Keywords: Others
[ Paper ]

Author Information

Damien Scieur (Samsung - SAIT AI Lab, Montreal)
Youngsung Kim (Samsung Advanced Institute of Technology)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors