Zero-Shot Kernel Learning

In this paper, we address an open problem of zero-shot learning. Its principle is based on learning a mapping that associates feature vectors extracted from i.e. images and attribute vectors that describe objects and/or scenes of interest. In turns, this allows classifying unseen object classes and/or scenes by matching feature vectors via mapping to a newly defined attribute vector describing a new class. Due to importance of such a learning task, there exist many methods that learn semantic, probabilistic, linear or piece-wise linear mappings. In contrast, we apply well-established kernel methods to learn a non-linear mapping between the feature and attribute spaces. We propose an easy learning objective inspired by the Linear Discriminant Analysis, Kernel-Target Alignment and Kernel Polarization methods [12, 8, 4] that promotes incoherence. We evaluate the performance of our algorithm on the Polynomial as well as shift-invariant Gaussian and Cauchy kernels. Despite simplicity of our approach, we obtain state-of-the-art results on several zero-shot learning datasets and benchmarks including a recent AWA2 dataset [45].

• Publication year

  2018

• Venue

  2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition

• Publication date

  2018-02-05

• Fields of study

  Mathematics, Computer Science

• Identifiers
  DOI 10.1109/CVPR.2018.00800arXiv 1802.01279
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• The proposed approach achieves state-of-the-art results on several zero-shot learning datasets and benchmarks, including the AWA2 dataset.

  Confidence 0.95

  All you need is Python (5d7gwfm5zu) extraction뀨 (7c402c1b98) reviewimjlk (vdp8mqzes2) reviewB (s683577b42) review
• An easy learning objective inspired by Linear Discriminant Analysis, Kernel-Target Alignment, and Kernel Polarization is proposed to promote incoherence.

  Confidence 0.95

  All you need is Python (5d7gwfm5zu) extraction뀨 (7c402c1b98) reviewimjlk (vdp8mqzes2) reviewB (s683577b42) review
• Kernel methods are applied to learn a non-linear mapping between the feature and attribute spaces for zero-shot learning.

  Confidence 0.95

  All you need is Python (5d7gwfm5zu) extraction뀨 (7c402c1b98) reviewimjlk (vdp8mqzes2) reviewB (s683577b42) review

• awa2 dataset

  dataset

  A recent zero-shot learning benchmark dataset used to evaluate the proposed algorithm.

  All you need is Python (5d7gwfm5zu) extraction뀨 (7c402c1b98) reviewimjlk (vdp8mqzes2) reviewB (s683577b42) review
• kernel methods

  method

  Well-established machine learning techniques applied here to learn a non-linear mapping between feature and attribute spaces.

  All you need is Python (5d7gwfm5zu) extraction뀨 (7c402c1b98) reviewimjlk (vdp8mqzes2) reviewB (s683577b42) review
• learning objective

  method

  An objective function inspired by Linear Discriminant Analysis, Kernel-Target Alignment, and Kernel Polarization designed to promote incoherence.

  All you need is Python (5d7gwfm5zu) extraction뀨 (7c402c1b98) reviewimjlk (vdp8mqzes2) reviewB (s683577b42) review
• non-linear mapping

  method

  A mapping function learned via kernel methods to bridge feature and attribute spaces.

  All you need is Python (5d7gwfm5zu) extraction뀨 (7c402c1b98) reviewimjlk (vdp8mqzes2) reviewB (s683577b42) review
• zero-shot learning

  task

  A learning task focused on classifying unseen object classes or scenes by mapping feature vectors to attribute vectors.

  All you need is Python (5d7gwfm5zu) extraction뀨 (7c402c1b98) reviewimjlk (vdp8mqzes2) reviewB (s683577b42) review

• Pattern Classification

  2021cited by this paper
• Geometry Aware Constrained Optimization Techniques for Deep Learning

  2018cited by this paper
• Semantic Autoencoder for Zero-Shot Learning

  2017cited by this paper
• Zero-Shot Learning — The Good, the Bad and the Ugly

  2017cited by this paper
• Higher-Order Occurrence Pooling for Bags-of-Words: Visual Concept Detection

  2017cited by this paper
• Zero-Shot Learning—A Comprehensive Evaluation of the Good, the Bad and the Ugly

  2017cited by this paper
• Joint Dimensionality Reduction and Metric Learning: A Geometric Take

  2017cited by this paper
• Domain Adaptation by Mixture of Alignments of Second-or Higher-Order Scatter Tensors

  2016cited by this paper
• Multi-cue Zero-Shot Learning with Strong Supervision

  2016cited by this paper
• Synthesized Classifiers for Zero-Shot Learning

  2016cited by this paper
• Latent Embeddings for Zero-Shot Classification

  2016influential reference
• Tensor Representations via Kernel Linearization for Action Recognition from 3D Skeletons

  2016cited by this paper
• Sparse Coding for Third-Order Super-Symmetric Tensor Descriptors with Application to Texture Recognition

  2016cited by this paper
• Deep Residual Learning for Image Recognition

  2015cited by this paper
• Zero-Shot Learning via Semantic Similarity Embedding

  2015cited by this paper
• An embarrassingly simple approach to zero-shot learning

  2015cited by this paper
• Evaluation of output embeddings for fine-grained image classification

  2014influential reference
• ImageNet Large Scale Visual Recognition Challenge

  2014cited by this paper
• GloVe: Global Vectors for Word Representation

  2014cited by this paper
• Texture Recognition

  2014cited by this paper
• Zero-Shot Learning by Convex Combination of Semantic Embeddings

  2013cited by this paper
• Label-Embedding for Attribute-Based Classification

  2013influential reference
• Distributed Representations of Words and Phrases and their Compositionality

  2013cited by this paper
• Zero-Shot Learning Through Cross-Modal Transfer

  2013influential reference
• DeViSE: A Deep Visual-Semantic Embedding Model

  2013influential reference
• Pattern Classification

  2012influential reference
• Large scale metric learning from equivalence constraints

  2012cited by this paper
• SUN attribute database: Discovering, annotating, and recognizing scene attributes

  2012cited by this paper
• Evaluating knowledge transfer and zero-shot learning in a large-scale setting

  2011cited by this paper
• Classification and clustering via dictionary learning with structured incoherence and shared features

  2010cited by this paper
• Caltech-UCSD Birds 200

  2010cited by this paper
• Safety in numbers: Learning categories from few examples with multi model knowledge transfer

  2010cited by this paper
• Describing objects by their attributes

  2009cited by this paper
• A least square kernel machine with box constraints

  2008cited by this paper
• Zero-data Learning of New Tasks

  2008cited by this paper
• Et al

  2008influential reference
• The Pascal Visual Object Classes Challenge 2006 ( VOC 2006 ) Results

  2006cited by this paper
• The PASCAL Visual Object Classes Challenge

  2006cited by this paper
• Learning by Kernel Polarization

  2005influential reference
• Large Margin Methods for Structured and Interdependent Output Variables

  2005cited by this paper
• Optimizing search engines using clickthrough data

  2002cited by this paper
• Learning with Kernels: support vector machines, regularization, optimization, and beyond

  2001influential reference
• On Kernel-Target Alignment

  2001influential reference
• Comparing support vector machines with Gaussian kernels to radial basis function classifiers

  1997cited by this paper
• THE USE OF MULTIPLE MEASUREMENTS IN TAXONOMIC PROBLEMS

  1936cited by this paper
• and as an in

  year unknowncited by this paper

• A Zero-shot Learning Method Based on Large Language Models for Multi-modal Knowledge Graph Embedding

  2025cites this paper
• Robust Generalized Zero-Shot Learning via Dual-Stream Variational Autoencoders and Out-of-Distribution Detection

  2025cites this paper
• Saliency-guided meta-hallucinator for few-shot learning

  2024cites this paper
• Knowledge Graph Enhancement for Fine-Grained Zero-Shot Learning on ImageNet21K

  2024cites this paper
• Sustainable and Transferable Traffic Sign Recognition for Intelligent Transportation Systems

  2023cites this paper
• Compound Projection Learning for Bridging Seen and Unseen Objects

  2023cites this paper
• Prototypical Model with Information-Theoretic Loss Functions for Generalized Zero-Shot Learning

  2023influential citation
• No Adversaries to Zero-Shot Learning: Distilling an Ensemble of Gaussian Feature Generators

  2023influential citation
• A review on multimodal zero‐shot learning

  2023cites this paper
• Towards Zero-Shot Learning: A Brief Review and an Attention-Based Embedding Network

  2023cites this paper
• Zero Shot Emoji Prediction using Multimodal Emoji Embeddings

  2022cites this paper
• Zero shot image classification system using an optimized generalized adversarial network

  2022cites this paper
• Semantic Enhanced Knowledge Graph for Large-Scale Zero-Shot Learning

  2022cites this paper
• Kernel Relative-prototype Spectral Filtering for Few-shot Learning

  2022cites this paper
• Learning exclusive discriminative semantic information for zero-shot learning

  2022cites this paper
• SimpleZSL: Extremely Simple and Fast Zero-Shot Learning with Nearest Neighbor Classifiers

  2022cites this paper
• A Zero-shot Learning Method with a Multi-Modal Knowledge Graph

  2022cites this paper
• Refinement Matters: Textual Description Needs to be Refined for Zero-shot Learning

  2022cites this paper
• Trading-off Information Modalities in Zero-shot Classification

  2022influential citation
• Hybrid Routing Transformer for Zero-Shot Learning

  2022cites this paper
• Zero-Shot Embedding via Regularization-Based Recollection and Residual Familiarity Processes

  2022cites this paper
• Zero-shot learning via a specific rank-controlled semantic autoencoder

  2022cites this paper
• Semantically Grounded Visual Embeddings for Zero-Shot Learning

  2022cites this paper
• Attribute-Modulated Generative Meta Learning for Zero-Shot Classification

  2021cites this paper
• Aberystwyth University Relation-based Discriminative Cooperation Network for Zero-Shot Classification

  2021cites this paper
• Aberystwyth University Relation-based Discriminative Cooperation Network for Zero-Shot Classification

  2021cites this paper
• Dual Progressive Prototype Network for Generalized Zero-Shot Learning

  2021cites this paper
• Pho(SC)-CTC—a hybrid approach towards zero-shot word image recognition

  2021cites this paper
• Double Discriminative Graph Regularized Semantic Auto-Encoder for Zero-shot Learning

  2021cites this paper
• A Joint Model of Adaptive Clustering and Multi-kernel Learning for Entity Alignment

  2021influential citation
• A plug-in attribute correction module for generalized zero-shot learning

  2021cites this paper
• Zero-Shot Learning via Discriminative Dual Semantic Auto-Encoder

  2021cites this paper
• Semi-Supervised Low-Rank Semantics Grouping for Zero-Shot Learning

  2021cites this paper
• Generalized Zero-Shot Learning With Multiple Graph Adaptive Generative Networks

  2021cites this paper
• Task Aligned Generative Meta-learning for Zero-shot Learning

  2021cites this paper
• VMAN: A Virtual Mainstay Alignment Network for Transductive Zero-Shot Learning

  2021cites this paper
• Dual discriminative auto-encoder network for zero shot image recognition

  2021cites this paper
• Attribute-Modulated Generative Meta Learning for Zero-Shot Learning

  2021cites this paper
• Recognizing Bengali Word Images - A Zero-Shot Learning Perspective

  2021cites this paper
• Zero-Shot Image Classification Based on a Learnable Deep Metric

  2021cites this paper
• Zero-Shot Learning for IMU-Based Activity Recognition Using Video Embeddings

  2021cites this paper
• Pho(SC)Net: An Approach Towards Zero-shot Word Image Recognition in Historical Documents

  2021cites this paper
• Cross-Modal Representation Reconstruction for Zero-Shot Classification

  2021cites this paper
• Near-Real Feature Generative Network for Generalized Zero-Shot Learning

  2021cites this paper
• Consistent Representation Learning Across Modalities for Zero-Shot Image Recognition

  2021cites this paper
• Prototypical Model with Novel Information-theoretic Loss Function for Generalized Zero Shot Learning

  2021cites this paper
• Meta-Learning for Multi-Label Few-Shot Classification

  2021cites this paper
• Kernelized distance learning for zero-shot recognition

  2021cites this paper
• SDM-Net: A simple and effective model for generalized zero-shot learning

  2021influential citation
• TranScreen: Transfer Learning on Graph-Based Anti-Cancer Virtual Screening Model

  2020cites this paper
• Label-activating framework for zero-shot learning

  2020cites this paper
• Rethinking Class Relations: Absolute-relative Few-shot Learning

  2020cites this paper
• Few-shot Action Recognition via Improved Attention with Self-supervision

  2020cites this paper
• Multi-Modality Adversarial Auto-Encoder for Zero-Shot Learning

  2020cites this paper
• From Anchor Generation to Distribution Alignment: Learning a Discriminative Embedding Space for Zero-Shot Recognition

  2020cites this paper
• Model-Agnostic Metric for Zero-Shot Learning

  2020cites this paper
• Guided CNN for generalized zero-shot and open-set recognition using visual and semantic prototypes

  2020cites this paper
• Meta-Learning for Generalized Zero-Shot Learning

  2020cites this paper
• Long-tail learning with attributes

  2020cites this paper
• A Joint Label Space for Generalized Zero-Shot Classification

  2020cites this paper
• Learning discriminative domain-invariant prototypes for generalized zero shot learning

  2020cites this paper
• Pseudo distribution on unseen classes for generalized zero shot learning

  2020cites this paper
• Zero-Shot Learning with Common Sense Knowledge Graphs

  2020cites this paper
• Invertible Zero-Shot Recognition Flows

  2020cites this paper
• Cooperative Coupled Generative Networks for Generalized Zero-Shot Learning

  2020cites this paper
• Enhancing Generalized Zero-Shot Learning via Adversarial Visual-Semantic Interaction

  2020cites this paper
• A Simple Discriminative Dual Semantic Auto-encoder for Zero-shot Classification

  2020cites this paper
• A Boundary Based Out-of-Distribution Classifier for Generalized Zero-Shot Learning

  2020cites this paper
• Prior Knowledge about Attributes: Learning a More Effective Potential Space for Zero-Shot Recognition

  2020cites this paper
• Learning Discriminative Projection With Visual Semantic Alignment for Generalized Zero Shot Learning

  2020cites this paper
• Prototype Relaxation With Robust Principal Component Analysis for Zero Shot Learning

  2020cites this paper
• From generalized zero-shot learning to long-tail with class descriptors

  2020cites this paper
• Two-Level Adversarial Visual-Semantic Coupling for Generalized Zero-shot Learning

  2020cites this paper
• Few-Shot Zero-Shot Learning: Knowledge Transfer with Less Supervision

  2020influential citation
• Bidirectional Mapping Coupled GAN for Generalized Zero-Shot Learning

  2020cites this paper
• Rethinking Class Relations: Absolute-relative Supervised and Unsupervised Few-shot Learning

  2020cites this paper
• Deep Zero- and Few-shot Learning in Computer Vision

  2020cites this paper
• Zero-Shot Learning via Category-Specific Visual-Semantic Mapping and Label Refinement

  2019cites this paper
• A Probabilistic Zero-Shot Learning Method via Latent Nonnegative Prototype Synthesis of Unseen Classes

  2019cites this paper
• Cross-Modal Mapping for Generalized Zero-Shot Learning by Soft-Labeling

  2019cites this paper
• Cross-Layer Autoencoder for Zero-Shot Learning

  2019cites this paper
• Relation-based Generalized Zero-shot Classification with the Domain Discriminator on the shared representation

  2019cites this paper
• An Optimal Transport Framework for Zero-Shot Learning

  2019influential citation
• Marginalized Latent Semantic Encoder for Zero-Shot Learning

  2019influential citation
• Semantic Similarity Based Softmax Classifier for Zero-Shot Learning

  2019cites this paper
• A Meta-Learning Framework for Generalized Zero-Shot Learning

  2019cites this paper
• Visual and Semantic Prototypes-Jointly Guided CNN for Generalized Zero-shot Learning

  2019cites this paper
• Domain-Specific Embedding Network for Zero-Shot Recognition

  2019cites this paper
• Zero-Shot Learning with Few Seen Class Samples

  2019cites this paper
• Graph and Autoencoder Based Feature Extraction for Zero-shot Learning

  2019cites this paper
• Landmark Selection for Zero-shot Learning

  2019influential citation
• Generalized Zero Shot Learning via Synthesis Pseudo Features

  2019cites this paper
• Supplementary Material : Min-Max Statistical Alignment for Transfer Learning

  2019cites this paper
• Min-Max Statistical Alignment for Transfer Learning

  2019cites this paper
• Discriminative Embedding Autoencoder With a Regressor Feedback for Zero-Shot Learning

  2019cites this paper
• Dual Adversarial Semantics-Consistent Network for Generalized Zero-Shot Learning

  2019cites this paper
• Few-Shot Learning via Saliency-Guided Hallucination of Samples

  2019cites this paper
• Zero Shot Learning with the Isoperimetric Loss

  2019influential citation
• Zero-Shot Learning Via Recurrent Knowledge Transfer

  2019cites this paper
• Dual-View Ranking with Hardness Assessment for Zero-Shot Learning

  2019cites this paper