Transductive Multi-View Zero-Shot Learning
Open accessThe DOI matches its Crossref record; 72 of the 74 references it cites were found in a registry or an index.
Abstract
Most existing zero-shot learning approaches exploit transfer learning via an intermediate semantic representation shared between an annotated auxiliary dataset and a target dataset with different classes and no annotation. A projection from a low-level feature space to the semantic representation space is learned from the auxiliary dataset and applied without adaptation to the target dataset. In this paper we identify two inherent limitations with these approaches. First, due to having disjoint and potentially unrelated classes, the projection functions learned from the auxiliary dataset/domain are biased when applied directly to the target dataset/domain. We call this problem the projection domain shift problem and propose a novel framework, transductive multi-view embedding, to solve it. The second limitation is the prototype sparsity problem which refers to the fact that for each target class, only a single prototype is available for zero-shot learning given a semantic representation. To overcome this problem, a novel heterogeneous multi-view hypergraph label propagation method is formulated for zero-shot learning in the transductive embedding space. It effectively exploits the complementary information offered by different semantic representations and takes advantage of the manifold structures of multiple representation spaces in a coherent manner. We demonstrate through extensive experiments that the proposed approach (1) rectifies the projection shift between the auxiliary and target domains, (2) exploits the complementarity of multiple semantic representations, (3) significantly outperforms existing methods for both zero-shot and N-shot recognition on three image and video benchmark datasets, and (4) enables novel cross-view annotation tasks.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155 · cited 1×
“What is interesting to see here is that Overfeat > Decaf since both are based on the same Convolutional Neural Network (CNN) model of [32].”
BackgroundCollective dynamics of ‘small-world’ networks
Watts, D. J., & Strogatz, S. H. (1998) · Nature · doi:10.1038/30918 · Cited by 43,060
LIBSVM: A library for support vector machines
Chang, C.-C., & Lin, C.-J. (2011) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/1961189.1961199 · Cited by 40,918 · cited 1×
“However, this does not effectively exploit the multi-view complementarity, and suffers from labelled data (prototype) sparsity.”
Background(2008) · Journal of Machine Learning Research · Cited by 33,051
Pan, S. J., & Yang, Q. (2010) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2009.191 · Cited by 23,387
Learning Deep Architectures for AI
Bengio, Y. (2009) · Foundations and Trends® in Machine Learning · doi:10.1561/2200000006 · Cited by 9,239 · cited 1×
“The advantages of heterogeneous hypergraphs We argue that the pairwise similarity of heterogeneous hypergraph is a distributed representation [3].”
BackgroundRecognition-by-components: A theory of human image understanding.
Biederman, I. (1987) · Psychological Review · doi:10.1037/0033-295x.94.2.115 · Cited by 5,621 · cited 1×
“Humans can distinguish 30,000 basic object classes [3] and many more subordinate ones (e.g. breeds of dogs).”
BackgroundLearning Deep Architectures for AI
Bengio, Y. (2009) · now publishers, Inc. eBooks · doi:10.1561/9781601982957 · Cited by 4,941
Semi-Supervised Learning Literature Survey
Zhu, X. (2005) · Minds at UW (University of Wisconsin) · Cited by 4,505
DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition
Donahue, J., Jia, Y., Vinyals, O., Hoffman, J., Zhang, N., Tzeng, E., & Darrell, T. (2013) · arXiv · doi:10.48550/arxiv.1310.1531 · Cited by 4,019 · cited 1×
“In addition, the Decaf (D) [10] feature is also considered for AwA.”
Background
Research works printed but not linked: 5
- [39] E. Rosch. Classification of real-world objects: Origins and representations in cognition. Thinking: Readings in Cognitive Science , 1977. -
- [32] T. Mikolov, K. Chen, G. Corrado, and J. Dean. Efficient estimation of word representation in vector space. In Proceedings of Workshop at ICLR , 2013. -
- [33] M. Palatucci, G. Hinton, D. Pomerleau, and T. M. Mitchell. Zeroshot learning with semantic output codes. In NIPS , 2009. -
- [34] D. Parikh and K. Grauman. Relative attributes. In ICCV , 2011. -
- [21] Y. Huang, Q. Liu, and D. Metaxas. Video object segmentation by hypergraph cut. In CVPR , 2009. -
Cited by
Items 501–512 of 512
Uncertainty-Aware Zero-Shot Learning via Evidential Regression and Weighted Semantic SimilarityPreprint
She, B., Ye, Z., Liu, D., Geng, G., Jin, X., & Chen, Y. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6365352
Will It Zero-Shot?:Predicting Zero-Shot Classification Performance for Arbitrary Queries
Robbins, K., Liu, X., Wu, Y., Sun, L., McPeak, G., Stylianou, A., & Pless, R. B. (2025) · IEEE ... International Conference on Data Mining workshops · doi:10.1109/icdmw69685.2025.00181
“Early efforts to solve this problem involved the identification of attribute vectors and the composition of attributes by linear combinations of these vectors, as demonstrated in studies [4]– [6].”
Liu, Z., Ren, Y., Zhu, Z., Li, T., Xu, G., & Wu, X. (2026) · Journal of Visualization · doi:10.1007/s12650-026-01113-9
Dual Adversarial Perturbations for Zero-Shot Learning
Wang, L., Chen, S., Xie, G.-S., Yu, C., You, X., Peng, Q., & You, X. (2026) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2026.3679600
“Transductive approaches directly incorporate unlabeled unseen data during training to reduce the domain gap [34], [35], [36].”
Deep Multimodal Learning for Joint Textual and Visual Reasoning
Bordes, P. (2020) · theses.fr (ABES) · doi:10.70675/8bd1beefz5760z4f96z8d8cz87e025608c32
Dual-Path Conditional Diffusion Model With Attribute Consistency for Zero-Shot Fault Diagnosis
Liao, W., Wu, L., Xu, S., & Fujimura, S. (2026) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2026.3683414
A Contemporary Analysis of Zero-Shot Learning in Machine Vision Tasks
Arun, V. S., & Simon, P. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-18135-0_15
Generative Zero-Shot Hashing for Multi-Label Image Retrieval
Ge, M., Qin, Q., Zhang, W., & Huang, L. (2026) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3815114
“Based on the methods of obtaining instances, zero-shot learning can be categorized into three groups [65]: compatibility methods [54, 69], transductive methods [13, 47, 52], and synthesizing methods [5, 50].”
Generalized Zero-Shot Learning for Evolving Network Device Identification
Wang, Z., Jin, M., Tang, Z., Chen, D., Wei, X., & You, L. (2026) · Electronics · doi:10.3390/electronics15112320
侯梦涛, Xu, C., & Yang, H. (2026) · Measurement Science and Technology · doi:10.1088/1361-6501/aea7bf
“Since the optimization is confined to the discriminative patterns of observed data, the learned projection inevitably develops a bias toward seen classes [21].”
Multi-View Clustering via Low-Rank and Smooth Tensorized Label Learning
Liu, Y., Gao, Q., Duan, Y., & Xue, X. (2026) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3767308.3836411
Metadata
- Authors
- Fu, Y., Hospedales, T. M., Xiang, T., & Gong, S.
- Affiliations
- Walt Disney (United States) (US)
- Queen Mary University of London (GB)
- Venue
- IEEE Transactions on Pattern Analysis and Machine Intelligence
- Published
- 2015-11-01
- Volume
- 37
- Issue
- 11
- Pages
- 2332-2345
- ISSN
- 0162-8828 (ISSN-L)
- 2160-9292
- OpenAlex
- W2141350700
- Cited by
- 514
- References
- 74
- Citations by year
- 2026: 17 · 2025: 32 · 2024: 32 · 2023: 52 · 2022: 60 · 2021: 75 · 2020: 62 · 2019: 57 · 2018: 64 · 2017: 41 · 2016: 21 · 2015: 1
- FWCI
- 41.51
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE Transactions on Pattern Analysis and Machine Intelligence (publishedVersion)
- PubMed (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- Queen Mary Research Online (Queen Mary University of London) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
- Keywords
- zero-shot learning, transductive learning, multi-view embedding, semantic representation, transfer learning, manifold structure
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.