Synthesizing Samples fro Zero-shot Learning

Guo, Yuchen and Ding, Guiguang and Han, Jungong and Gao, Yue (2017) Synthesizing Samples fro Zero-shot Learning. In: Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence :. IJCAI, Melbourne, pp. 1774-1780. ISBN 9780999241103

[thumbnail of ijcai2017_submission_yuchenzeroshot]
PDF (ijcai2017_submission_yuchenzeroshot)
ijcai2017_submission_yuchenzeroshot.pdf - Accepted Version
Available under License Creative Commons Attribution-NonCommercial.

Download (1MB)


Zero-shot learning (ZSL) is to construct recognition models for unseen target classes that have no labeled samples for training. It utilizes the class attributes or semantic vectors as side information and transfers supervision information from related source classes with abundant labeled samples. Existing ZSL approaches adopt an intermediary embedding space to measure the similarity between a sample and the attributes of a target class to perform zero-shot classification. However, this way may suffer from the information loss caused by the embedding process and the similarity measure cannot fully make use of the data distribution. In this paper, we propose a novel approach which turns the ZSL problem into a conventional supervised learning problem by synthesizing samples for the unseen classes. Firstly, the probability distribution of an unseen class is estimated by using the knowledge from seen classes and the class attributes. Secondly, the samples are synthesized based on the distribution for the unseen class. Finally, we can train any supervised classifiers based on the synthesized samples. Extensive experiments on benchmarks demonstrate the superiority of the proposed approach to the state-of-the-art ZSL approaches.

Item Type:
Contribution in Book/Report/Proceedings
ID Code:
Deposited By:
Deposited On:
17 Oct 2017 08:26
Last Modified:
03 May 2024 23:22