Attribute-Based Synthetic Network (ABS-Net): Learning more from pseudo feature representations

作者：

Highlights：

• ABS-Net is proposed to deal with the lack of annotated samples.

• ABS-Net fills the gaps between seen and unseen concepts.

• ABS-Net achieves competitive results on some zero-shot learning benchmark datasets.

• ABS-Net realizes feature-level data augmentation for supervised learning.

• Experiments on C-MNIST demonstrate the effectiveness of the proposed model.

摘要

•ABS-Net is proposed to deal with the lack of annotated samples.•ABS-Net fills the gaps between seen and unseen concepts.•ABS-Net achieves competitive results on some zero-shot learning benchmark datasets.•ABS-Net realizes feature-level data augmentation for supervised learning.•Experiments on C-MNIST demonstrate the effectiveness of the proposed model.

论文关键词：Pseudo feature representation,Zero-shot learning,Supervised learning,Data augmentation,Attribute learning

论文评审过程：Received 20 September 2017, Revised 30 November 2017, Accepted 4 March 2018, Available online 6 March 2018, Version of Record 19 March 2018.

论文官网地址：https://doi.org/10.1016/j.patcog.2018.03.006