Abstract

Most existing Zero-Shot Learning (ZSL) methods have the strong bias problem, in which instances of unseen (target) classes tend to be categorized as one of the seen (source) classes. So they yield poor performance after being deployed in the generalized ZSL settings. In this paper, we propose a straightforward yet effective method named Quasi-Fully Supervised Learning (QFSL) to alleviate the bias problem. Our method follows the way of transductive learning, which assumes that both the labeled source images and unlabeled target images are available for training. In the semantic embedding space, the labeled source images are mapped to several fixed points specified by the source categories, and the unlabeled target images are forced to be mapped to other points specified by the target categories. Experiments conducted on AwA2, CUB and SUN datasets demonstrate that our method outperforms existing state-ofthe-art approaches by a huge margin of 9.3 ∼ 24.5% following generalized ZSL settings, and by a large margin of 0.2 ∼ 16.2% following conventional ZSL settings.

https://arxiv.org/pdf/1803.11320.pdf

Jie Song1 , Chengchao Shen1 , Yezhou Yang2 , Yang Liu3 , and Mingli Song1 1College of Computer Science and Technology, Zhejiang University, Hangzhou, China 2Arizona State University, Tempe, USA 3Alibaba Group, Hangzhou, China

 

What learned from this work:

I am a new guy to this topic, but I am preparing myself to get into it now.

•  The key point of this work is to propose a sort of new sub-topic named Quasi-Fully Supervised Learning, QFSL.   Most previous works are constrained in the Conventional settings, which assumes that all test images are from target classes (unlabeled), however,  this work changes this assumption to be Generalized settings, which is test images should be from both target and source classes.
• One concern or thought is what kind of answers it would give when most test images are not belonging to any class in both target and source.   Could the issue collapse this model? How can we have the model to learn how to learn?

The following points only belong to me:

• The background of Zero-shot learning (ZSL)
• The difference between Inductive ZSL and Transductive ZSL
•  The Strong bias issue
•  The interesting architecture of the proposed QFSL model

 

 

 

 

 

 

 

A Chinese blog also shares a well-written interpretation about this work: https://mp.weixin.qq.com/s/od9i5Pf8-E0ouoih6Z97fQ