Classifier and Exemplar Synthesis for Zero-Shot Learning

By Soravit Changpinyo, Wei-Lun Chao, Boqing Gong, Fei Sha. 2018

52 citations recorded by the source corpus.

Zero-shot learning (ZSL) enables solving a task without the need to see its examples. In this paper, we propose two ZSL frameworks that learn to synthesize parameters for novel unseen classes. First, we propose to cast the problem of ZSL as learning manifold embeddings from graphs composed of object classes, leading to a flexible approach that synthesizes “classifiers” for the unseen classes. Then, we define an auxiliary task of synthesizing “exemplars” for the unseen classes to be used as an automatic denoising mechanism for any existing ZSL approaches or as an effective ZSL model by itself. On five visual recognition benchmark datasets, we demonstrate the superior performances of our proposed frameworks in various scenarios of both conventional and generalized ZSL. Finally, we provide valuable insights through a series of empirical analyses, among which are a comparison of semantic representations on the full ImageNet benchmark as well as a comparison of metrics used in generalized ZSL. Our code and data are publicly available at https://github.com/pujols/Zero-shot-learning-journal.

Subfields: Artificial Intelligence

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Han-Jia Ye
2. De-chuan Zhan
3. Fei Sha
4. Hexiang Hu
5. Kilian Q. Weinberger
6. Felix Wu
7. Da-Wei Zhou
8. Yejin Choi

Citations in the corpus

• Simplifying Graph Convolutional Networks (2019, background): “Again, this downstream application demonstrates that learned graph convolution filters are superfluous; similar to Changpinyo et al. (2018)’s observation that GCNs may not be necessary.”
• Learning Adaptive Classifiers Synthesis for Generalized Few-Shot Learning (2019, background): “Zero-shot learning (ZSL) (Lampert et al. 2014;Akata et al. 2013; Xian et al. 2017; Changpinyo et al. 2020) is a popular idea for addressing learningwithout labeled data.”
• Learning Classifier Synthesis for Generalized Few-Shot Learning (2019, methodology): “Our model is based on classifier synthesis [4, 6] with a learned neural dictionary.”
• A Review of Generalized Zero-Shot Learning Methods (2020, methodology): “Later, several studies used the calibrated stacking technique to solve the GZSL problem [73], [78], [79], [80], [81].”