Deep Uniformly Distributed Centers on a Hypersphere for Open Set Recognition

Hakan Cevikalp, Hasan Serhan Yavuz, Hasan Saribas
Proceedings of the 15th Asian Conference on Machine Learning, PMLR 222:217-230, 2024.

Abstract

This study introduces a new approach for open set recognition, wherein we propose a novel method utilizing uniformly distributed centers on a hypersphere. Each class in the proposed method is represented by a center, and these centers and features of the deep learning architecture are jointly learned from the training data in an end-to-end fashion. We ensure that the centers lie on the boundary of a hypersphere whose center is positioned at the origin. The class-specific samples are compelled by the proposed loss function to be closer to their respective centers. In open set recognition scenarios, an additional loss term is employed to separate the background samples from the known class centers. The assignment of test samples to classes is based on the Euclidean distances calculated from the learned class centers. Experimental results show that the proposed method yields the state-of-the-art accuracies on open set recognition datasets.

Cite this Paper

BibTeX
@InProceedings{pmlr-v222-cevikalp24a, title = {Deep Uniformly Distributed Centers on a Hypersphere for Open Set Recognition}, author = {Cevikalp, Hakan and Yavuz, Hasan Serhan and Saribas, Hasan}, booktitle = {Proceedings of the 15th Asian Conference on Machine Learning}, pages = {217--230}, year = {2024}, editor = {Yanıkoğlu, Berrin and Buntine, Wray}, volume = {222}, series = {Proceedings of Machine Learning Research}, month = {11--14 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v222/cevikalp24a/cevikalp24a.pdf}, url = {https://proceedings.mlr.press/v222/cevikalp24a.html}, abstract = {This study introduces a new approach for open set recognition, wherein we propose a novel method utilizing uniformly distributed centers on a hypersphere. Each class in the proposed method is represented by a center, and these centers and features of the deep learning architecture are jointly learned from the training data in an end-to-end fashion. We ensure that the centers lie on the boundary of a hypersphere whose center is positioned at the origin. The class-specific samples are compelled by the proposed loss function to be closer to their respective centers. In open set recognition scenarios, an additional loss term is employed to separate the background samples from the known class centers. The assignment of test samples to classes is based on the Euclidean distances calculated from the learned class centers. Experimental results show that the proposed method yields the state-of-the-art accuracies on open set recognition datasets.} }
Endnote
%0 Conference Paper %T Deep Uniformly Distributed Centers on a Hypersphere for Open Set Recognition %A Hakan Cevikalp %A Hasan Serhan Yavuz %A Hasan Saribas %B Proceedings of the 15th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Berrin Yanıkoğlu %E Wray Buntine %F pmlr-v222-cevikalp24a %I PMLR %P 217--230 %U https://proceedings.mlr.press/v222/cevikalp24a.html %V 222 %X This study introduces a new approach for open set recognition, wherein we propose a novel method utilizing uniformly distributed centers on a hypersphere. Each class in the proposed method is represented by a center, and these centers and features of the deep learning architecture are jointly learned from the training data in an end-to-end fashion. We ensure that the centers lie on the boundary of a hypersphere whose center is positioned at the origin. The class-specific samples are compelled by the proposed loss function to be closer to their respective centers. In open set recognition scenarios, an additional loss term is employed to separate the background samples from the known class centers. The assignment of test samples to classes is based on the Euclidean distances calculated from the learned class centers. Experimental results show that the proposed method yields the state-of-the-art accuracies on open set recognition datasets.
APA
Cevikalp, H., Yavuz, H.S. & Saribas, H.. (2024). Deep Uniformly Distributed Centers on a Hypersphere for Open Set Recognition. Proceedings of the 15th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 222:217-230 Available from https://proceedings.mlr.press/v222/cevikalp24a.html.

Related Material