Cross Knowledge-based Generative Zero-Shot Learning Approach with Taxonomy Regularization

@article{Xie2021CrossKG,
  title={Cross Knowledge-based Generative Zero-Shot Learning Approach with Taxonomy Regularization},
  author={Cheng Xie and Hongxin Xiang and Ting Zeng and Yun Yang and Beibei Yu and Qing Liu},
  journal={Neural networks : the official journal of the International Neural Network Society},
  year={2021},
  volume={139},
  pages={
          168-178
        }
}
  • Cheng Xie, Hongxin Xiang, Qing Liu
  • Published 25 January 2021
  • Computer Science
  • Neural networks : the official journal of the International Neural Network Society

Figures and Tables from this paper

Multi-Knowledge Fusion Network for Generalized Zero-Shot Learning
TLDR
This work proposes a novel generative ZSL method to learn more generalized features from multi-knowledge in semantic-to-visual embedding and shows that this approach consistently outperforms these state-of-the-art methods on a large number of available benchmarks on the generalized ZSL (GZSL).
Multi-Knowledge Fusion for New Feature Generation in Generalized Zero-Shot Learning
TLDR
This work proposes a novel generative ZSL method to learn more generalized features from multi-knowledge with continuously generated new semantics in semantic-to-visual embedding, and generates more generalized visual features by adaptively fusing visual features from different knowledge domain.
lass knowledge overlay to visual feature learning for zero-shot image lassification
TLDR
A novel zero-shot learning approach, GAN-CST, based on class knowledge to visual feature learning to tackle the problem of zero- shot image classification and tabulate results on a number of benchmark datasets demonstrating that the proposed model delivers superior performance over state-of-the-art approaches.
A Review of Generalized Zero-Shot Learning Methods
TLDR
This review paper presents a hierarchical categorization of the GZSL methods and discusses the representative methods of each category, and discusses several research directions for future studies.
A comprehensive survey of zero-shot image classification: methods, implementation, and fair evaluation
TLDR
In this article, a comprehensive survey is given on the methodology, implementation, and fair evaluations for practical and applied computing facets on the recent progress of zero-shot learning.
TLSCA-SVM Fault Diagnosis Optimization Method Based on Transfer Learning
TLDR
A novel fault-diagnosis-classification optimization method, namely TLSCA-SVM, which combines the sine cosine algorithm and support vector machine (SCA) with transfer learning with the aim of achieving the effect of transfer learning in the case of less failure data is proposed here.

References

SHOWING 1-10 OF 80 REFERENCES
Transductive Zero-Shot Learning With Adaptive Structural Embedding
TLDR
Two corresponding methods named Adaptive STructural Embedding (ASTE) and Self-PAced Selective Strategy (SPASS) for visual-semantic embedding and domain adaptation in cross-modality learning and unseen class prediction steps are presented.
Semantic Autoencoder for Zero-Shot Learning
TLDR
This work presents a novel solution to ZSL based on learning a Semantic AutoEncoder (SAE), which outperforms significantly the existing ZSL models with the additional benefit of lower computational cost and beats the state-of-the-art when the SAE is applied to supervised clustering problem.
Zero Shot Learning via Low-rank Embedded Semantic AutoEncoder
TLDR
A novel framework named Low-rank Embedded Semantic AutoEncoder (LESAE) is formulated to jointly seek a low-rank mapping to link visual features with their semantic representations to reconstruct the original data with the learned mapping.
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders
TLDR
This work proposes a model where a shared latent space of image features and class embeddings is learned by modality-specific aligned variational autoencoders, and align the distributions learned from images and from side-information to construct latent features that contain the essential multi-modal information associated with unseen classes.
Transductive Multi-View Zero-Shot Learning
TLDR
A novel heterogeneous multi-view hypergraph label propagation method is formulated for zero-shot learning in the transductive embedding space that rectifies the projection shift between the auxiliary and target domains, exploits the complementarity of multiple semantic representations, and significantly outperforms existing methods for both zero- shot and N-shot recognition.
Attentive Region Embedding Network for Zero-Shot Learning
  • Guosen Xie, Li Liu, L. Shao
  • Computer Science
    2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2019
TLDR
To discover (semantic) regions, the attentive region embedding network (AREN) is proposed, which is tailored to advance the ZSL task and achieves state-of-the-art performances under ZSLSetting, and compelling results under generalized ZSL setting.
Marginalized Latent Semantic Encoder for Zero-Shot Learning
  • Zhengming Ding, Hongfu Liu
  • Computer Science
    2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2019
TLDR
This paper designs a Marginalized Latent Semantic Encoder (MLSE), which is learned on the augmented seen visual features and the latent semantic representation, and whose latent semantics are discovered under an adaptive graph reconstruction scheme based on the provided semantics.
Gradient Matching Generative Networks for Zero-Shot Learning
TLDR
This work proposes a generative model that can naturally learn from unsupervised examples, and synthesize training examples for unseen classes purely based on their class embeddings, and therefore, reduce the zero-shot learning problem into a supervised classification task.
Feature Generating Networks for Zero-Shot Learning
TLDR
A novel generative adversarial network (GAN) that synthesizes CNN features conditioned on class-level semantic information, offering a shortcut directly from a semantic descriptor of a class to a class-conditional feature distribution.
...
...