Knowledge Distillation Classifier Generation Network for Zero-Shot Learning
- 29 September 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 34 (6) , 3183-3194
Abstract
In this article, we present a conceptually simple but effective framework called knowledge distillation classifier generation network (KDCGN) for zero-shot learning (ZSL), where the learning agent requires recognizing unseen classes that have no visual data for training. Different from the existing generative approaches that synthesize visual features for unseen classifiers' learning, the proposed framework directly generates classifiers for unseen classes conditioned on the corresponding class-level semantics. To ensure the generated classifiers to be discriminative to the visual features, we borrow the knowledge distillation idea to both supervise the classifier generation and distill the knowledge with, respectively, the visual classifiers and soft targets trained from a traditional classification network. Under this framework, we develop two, respectively, strategies, i.e., class augmentation and semantics guidance, to facilitate the supervision process from the perspectives of improving visual classifiers. Specifically, the class augmentation strategy incorporates some additional categories to train the visual classifiers, which regularizes the visual classifier weights to be compact, under supervision of which the generated classifiers will be more discriminative. The semantics-guidance strategy encodes the class semantics into the visual classifiers, which would facilitate the supervision process by minimizing the differences between the generated and the real-visual classifiers. To evaluate the effectiveness of the proposed framework, we have conducted extensive experiments on five datasets in image classification, i.e., AwA1, AwA2, CUB, FLO, and APY. Experimental results show that the proposed approach performs best in the traditional ZSL task and achieves a significant performance improvement on four out of the five datasets in the generalized ZSL task.Keywords
Funding Information
- National Natural Science Foundation of China (62002320, U19B2043, 61672456)
- Fundamental Research Funds for the Central Universities (2020QNA5010)
- Key Research and Development Program of Zhejiang Province, China (2021C01119)
This publication has 36 references indexed in Scilit:
- An Embarrassingly Simple Approach to Zero-Shot Learning2017
- An Empirical Study and Analysis of Generalized Zero-Shot Learning for Object Recognition in the Wild2016
- Synthesized Classifiers for Zero-Shot Learning2016
- Learning Deep Representations of Fine-Grained Visual Descriptions2016
- Latent Embeddings for Zero-Shot Classification2016
- Ridge Regression, Hubness, and Zero-Shot Learning2015
- Zero-shot object recognition by semantic manifold distance2015
- Evaluation of output embeddings for fine-grained image classification2015
- Building a bird recognition app and large scale dataset with citizen scientists: The fine print in fine-grained dataset collection2015
- An overview of statistical learning theoryIEEE Transactions on Neural Networks, 1999