Adaptive Graph Guided Embedding for Multi-label Annotation

  title={Adaptive Graph Guided Embedding for Multi-label Annotation},
  author={Lichen Wang and Zhengming Ding and Yun Raymond Fu},
Multi-label annotation is challenging since a large amount of well-labeled training data are required to achieve promising performance. However, providing such data is expensive while unlabeled data are widely available. To this end, we propose a novel Adaptive Graph Guided Embedding (AG2E) approach for multi-label annotation in a semi-supervised fashion, which utilizes limited labeled data associating with large-scale unlabeled data to facilitate learning performance. Specifically, a multi… 

Figures and Tables from this paper

Generic Multi-label Annotation via Adaptive Graph and Marginalized Augmentation

A generic multi-label learning framework based on Adaptive Graph and Marginalized Augmentation (AGMA) in a semi-supervised scenario and makes use of a small amount of labeled data associated with a lot of unlabeled data to boost the learning performance.

Adaptive Hypergraph Embedded Semi-Supervised Multi-Label Image Annotation

This work proposes a novel adaptive hypergraph learning (AHL) method for multilabel image annotation in a semisupervised way, in which both the limited labeled data and abundant unlabeled data are utilized to facilitate the annotation performance.

Dual Relation Semi-Supervised Multi-Label Learning

This work proposed a Dual Relation Semi-supervised Multi-label Learning (DRML) approach which jointly explores the feature distribution and the label relation simultaneously simultaneously and is an end-to-end model without any extra knowledge.

Generative Multi-Label Correlation Learning

A general and compact Multi-Label Correlation Learning (MUCO) framework that explicitly and effectively learns the latent label correlations by updating a label correlation tensor, which provides high accurate and interpretable prediction results.

Acknowledging the Unknown for Multi-label Learning with Single Positive Labels

This work proposes entropy-maximization (EM) loss to attain a special gradient regime for providing proper supervision signals and proposes asymmetric pseudo-labeling (APL), which adopts asymmetric-tolerance strategies and a self-paced procedure, to cooperate with EM loss and then provide more precise supervision.

Generative Correlation Discovery Network for Multi-label Learning

This paper proposes an end-to-end Generative Correlation Discovery Network (GCDN) method for multi-label learning that learns the label correlations based on a specifically-designed, simple but effective correlation discovery network to automatically discover the label correlation and considerately improve the label prediction accuracy.

PLMCL: Partial-Label Momentum Curriculum Learning for Multi-Label Image Classification

An end-to-end deep network, PLMCL (Partial-Label Momentum Curriculum Learning), that can learn to produce confident pseudo labels for both partially-labeled and unlabeled training images is proposed and outperforms many state-of-the-art multi-label classification methods under various partial-label settings on three different datasets.

Generic Embedded Semantic Dictionary for Robust Multi-Label Classification

This work proposes a Generic Embedded Semantic Dictionary (GESD) learning framework for robust multi-label classification, and explores a low-rank coding strategy to encode visual features with recovered label matrix by constructing an effective semantic dictionary.

Adaptive Graph Guided Disambiguation for Partial Label Learning

This work presents a unified framework which jointly optimizes the ground-truth labeling confidences, similarity graph and model parameters to achieve strong generalization performance and shows that PL-AGGD performs favorably against state-of-the-art partial label learning approaches.

Label Correlation Guided Deep Multi-View Image Annotation

An image annotation method by integrating deep multi-view latent space learning and label correlation guided image annotation into a unified framework, which is termed as Label Correlation guided Deep Multi-view image annotation (LCDM) method is proposed.



ML-MG: Multi-label Learning with Missing Labels Using a Mixed Graph

This work proposes a unified model of label dependencies by constructing a mixed graph, which jointly incorporates (i) instance-level similarity and class co-occurrence as undirected edges and (ii) semantic label hierarchy as directed edges.

An adaptive graph model for automatic image annotation

This paper proposes a novel automatic image annotation method based on manifold ranking learning, in which the visual and textual information are well integrated, and designs a new scheme named the Nearest Spanning Chain (NSC) to generate an adaptive similarity graph.

Low-Rank Embedded Ensemble Semantic Dictionary for Zero-Shot Learning

This work forms a novel framework to jointly seek a low-rank embedding and semantic dictionary to link visual features with their semantic representations, which manages to capture shared features across different observed classes.

Learning Transferable Subspace for Human Motion Segmentation

This work proposes a novel transferable subspace clustering approach by exploring useful information from relevant source data to enhance clustering performance in target temporal data and manages to transform the original data into a shared low-dimensional and distinctive feature space by jointly seeking an effective domain-invariant projection.

TagProp: Discriminative metric learning in nearest neighbor models for image auto-annotation

This work proposes TagProp, a discriminatively trained nearest neighbor model that allows the integration of metric learning by directly maximizing the log-likelihood of the tag predictions in the training set, and introduces a word specific sigmoidal modulation of the weighted neighbor tag predictions to boost the recall of rare words.

Beyond the point cloud: from transductive to semi-supervised learning

This paper constructs a family of data-dependent norms on Reproducing Kernel Hilbert Spaces (RKHS) that allow the structure of the RKHS to reflect the underlying geometry of the data.

Initialization Independent Clustering With Actively Self-Training Method

  • F. NieDong XuXuelong Li
  • Computer Science
    IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics)
  • 2012
An actively self-training clustering method, in which the samples are actively selected as training set to minimize an estimated Bayes error, and then explore semisupervised learning to perform clustering.

Zero-Shot Learning via Visual Abstraction

This paper proposes a new modality for ZSL using visual abstraction to learn difficult-to-describe concepts related to people and their interactions with others, and learns an explicit mapping between the abstract and real worlds.

On the Optimality of Classifier Chain for Multi-label Classification

This work first generalizes the CC model over a random label order, then presents a theoretical analysis of the generalization error for the proposed generalized model and proposes a dynamic programming based classifier chain (CC-DP) algorithm to search the globally optimal label order for CC and a greedy classifiers chain ( CC-Greedy) algorithms to find a locally optimal CC.

Unsupervised Feature Selection with Structured Graph Optimization

This work proposes an unsupervised feature selection approach which performs feature selection and local structure learning simultaneously, and constrain the similarity matrix to make it contain more accurate information of data structure, thus the proposed approach can select more valuable features.