arXiv Open Access 2017

Mimicking Ensemble Learning with Deep Branched Networks

Byungju Kim Youngsoo Kim Yeakang Lee Junmo Kim

Lihat Sumber

Abstrak

This paper proposes a branched residual network for image classification. It is known that high-level features of deep neural network are more representative than lower-level features. By sharing the low-level features, the network can allocate more memory to high-level features. The upper layers of our proposed network are branched, so that it mimics the ensemble learning. By mimicking ensemble learning with single network, we have achieved better performance on ImageNet classification task.

Topik & Kata Kunci

cs.CV

Penulis (4)

Byungju Kim

Youngsoo Kim

Yeakang Lee

Junmo Kim

Format Sitasi

APA MLA BibTeX

Kim, B., Kim, Y., Lee, Y., Kim, J. (2017). Mimicking Ensemble Learning with Deep Branched Networks. https://arxiv.org/abs/1702.06376

Akses Cepat

Lihat di Sumber

Informasi Jurnal

Tahun Terbit: 2017
Bahasa: en
Sumber Database: arXiv
Akses: Open Access ✓