This work focuses on the problem of multi-label learning with missing labels (MLML), which aims to label each test instance with multiple class labels given training instances that have an ...incomplete/partial set of these labels (i.e., some of their labels are missing). The key point to handle missing labels is propagating the label information from the provided labels to missing labels, through a dependency graph that each label of each instance is treated as a node. We build this graph by utilizing different types of label dependencies. Specifically, the instance-level similarity is served as undirected edges to connect the label nodes across different instances and the semantic label hierarchy is used as directed edges to connect different classes. This base graph is referred to as the mixed dependency graph, as it includes both undirected and directed edges. Furthermore, we present another two types of label dependencies to connect the label nodes across different classes. One is the class co-occurrence, which is also encoded as undirected edges. Combining with the above base graph, we obtain a new mixed graph, called mixed graph with co-occurrence (MG-CO). The other is the sparse and low rank decomposition of the whole label matrix, to embed high-order dependencies over all labels. Combining with the base graph, the new mixed graph is called as MG-SL (mixed graph with sparse and low rank decomposition). Based on MG-CO and MG-SL, we further propose two convex transductive formulations of the MLML problem, denoted as MLMG-CO and MLMG-SL respectively. In both formulations, the instance-level similarity is embedded through a quadratic smoothness term, while the semantic label hierarchy is used as a linear constraint. In MLMG-CO, the class co-occurrence is also formulated as a quadratic smoothness term, while the sparse and low rank decomposition is incorporated into MLMG-SL, through two additional matrices (one is assumed as sparse, and the other is assumed as low rank) and an equivalence constraint between the summation of this two matrices and the original label matrix. Interestingly, two important applications, including image annotation and tag based image retrieval, can be jointly handled using our proposed methods. Experimental results on several benchmark datasets show that our methods lead to significant improvements in performance and robustness to missing labels over the state-of-the-art methods.
Partial label learning aims to learn from training examples each associated with a set of
candidate
labels, among which only one label is valid for the training example. The basic strategy to learn ...from partial label examples is disambiguation, i.e. by trying to recover the ground-truth labeling information from the candidate label set. As one of the popular machine learning paradigms, maximum margin techniques have been employed to solve the partial label learning problem. Existing attempts perform disambiguation by optimizing the margin between the maximum modeling output from candidate labels and that from non-candidate ones. Nonetheless, this formulation ignores considering the margin between the ground-truth label and other candidate labels. In this paper, a new maximum margin formulation for partial label learning is proposed which directly optimizes the margin between the ground-truth label and all other labels. Specifically, the predictive model is learned via an alternating optimization procedure which coordinates the task of
ground-truth label identification
and
margin maximization
iteratively. Extensive experiments on artificial as well as real-world datasets show that the proposed approach is highly competitive to other well-established partial label learning approaches.
•We have proposed a fast algorithm for feature selection on the multi-label data.•Features that discriminate classes are linked to provide an undirected weighted graph.•Features relationships are ...defined based on correlation distance with labels.•PageRank algorithm ranks the features according to their importance in weighted graph.•The proposed multi-label graph based method outperforms competitive methods.
In multi-label data, each instance corresponds to a set of labels instead of one label whereby the instances belonging to a label in the corresponding column of that label are assigned 1, while instances that do not belong to that label are assigned 0 in the data set. This type of data is usually considered as high-dimensional data, so many methods, using machine learning algorithms, seek to choose the best subset of features for reducing the dimensionality of data and then to create an acceptable model for classification. In this paper, we have designed a fast algorithm for feature selection on the multi-label data using the PageRank algorithm, which is an effective method used to calculate the importance of web pages on the Internet. This algorithm, which is called multi-label graph-based feature selection (MGFS), first constructs an M × L matrix, called Correlation Distance Matrix (CDM), where M is the number of features and L represents the number of class labels. Then, MGFS creates a complete weighted graph, called Feature-Label Graph (FLG), where each feature is considered as a vertex, and the weight between two vertices (or features) represents their Euclidean distance in CDM. Finally, the importance of each graph vertex (or feature) is estimated via the PageRank algorithm. In the proposed method, the number of features can be determined by the user. To prove the performance of the proposed algorithm, we have tested this algorithm with several methods for multi-label feature selection and on several multi-label datasets with different dimensions. The results show the superiority of the proposed method in the classification criteria and run-time.
Feature selection used for dimensionality reduction of the feature space plays an important role in multi-label learning where high-dimensional data are involved. Although most existing multi-label ...feature selection approaches can deal with the problem of label ambiguity which mainly focuses on the assumption of uniform distribution with logical labels, it cannot be applied to many practical applications where the significance of related label for every instance tends to be different. To deal with this issue, in this study, label distribution learning covered with a certain real number of labels is introduced to design a model for the labeling-significance. Nevertheless, multi-label feature selection is limited to handling only labels consisting of logical relations. In order to solve this problem, combining the random variable distribution with granular computing, we first propose a label enhancement algorithm to transform logical labels in multi-label data into label distribution with more supervised information, which can mine the hidden label significance from every instance. On this basis, to remove some redundant or irrelevant features in multi-label data, a label distribution feature selection algorithm using mutual information and label enhancement is developed. Finally, the experimental results show that the performance of the proposed method is superior to the other state-of-the-art approaches when dealing with multi-label data.
When training a data-driven model, it is often assumed that the data distribution of the source domain and the target domain are the same.However, in the natural scenario, this assumption is usually ...not tenable, and it is easy to cause poor generalization ability of the model.Domain adaptation is a method proposed to improve the generalization ability of the model.It aligns the data distribution of the source domain and the target domain by learning the data characteristics of the two domains, so that the model trained in the source domain data can also perform well in the target domain with a small number of data labels or without data labels.In order to further improve the generalization ability of the model, existing researches have explored the know-ledge integrating into domain adaptive methods, which has high practical value.Firstly, we summarizes the development background of domain adaptive methods with knowledge integrating and the research status of related reviews.Then, the problem defi-nition and
Small-molecule fluorophores, such as fluorescein and rhodamine derivatives, are critical tools in modern biochemical and biological research. The field of chemical dyes is old; colored molecules were ...first discovered in the 1800s, and the fluorescein and rhodamine scaffolds have been known for over a century. Nevertheless, there has been a renaissance in using these dyes to create tools for biochemistry and biology. The application of modern chemistry, biochemistry, molecular genetics, and optical physics to these old structures enables and drives the development of novel, sophisticated fluorescent dyes. This critical review focuses on an important example of chemical biology-the melding of old and new chemical knowledge-leading to useful molecules for advanced biochemical and biological experiments.
Anti‐counterfeit labels protect many commercial goods, drugs, and currencies from counterfeiting activities. Recently, the designs of anti‐counterfeit labels have emerged utilizing surface‐enhanced ...Raman scattering (SERS) as a powerful technique, in which SERS‐active materials provide strong Raman signals of probe molecules. These signals are unique and, so far, have rarely been used in practical anti‐counterfeit labels applications. In this review, the general methodology of using Raman and SERS in designing anti‐counterfeit labels is first introduced. Then, two types of secret information in SERS labels, spectroscopic information and graphical information, are detailed and discussed with a focus on how the molecular information is encoded with SERS labels. Later, several advanced SERS labels are presented which combine existing security features such as barcode, quick response code, fluorescence, and unclonable features. Finally, the challenges in building usable SERS anti‐counterfeit labels are discussed and possible research directions are described.
Surface‐enhanced Raman scattering (SERS)‐based anti‐counterfeit labels are a new type of anti‐counterfeit label that has a high potential for future use in security applications. This review discusses various efforts and strategies in the development of SERS labels. These works enable the high novelty and high information capacity, which is of great importance to satisfy the need for security protection in the future.
Recently, label distribution learning (LDL) has drawn much attention in machine learning, where LDL model is learned from labelel instances. Different from single-label and multi-label annotations, ...label distributions describe the instance by multiple labels with different intensities and accommodate to more general scenes. Since most existing machine learning datasets merely provide logical labels, label distributions are unavailable in many real-world applications. To handle this problem, we propose two novel label enhancement methods, i.e., Label Enhancement with Sample Correlations (LESC) and generalized Label Enhancement with Sample Correlations (gLESC). More specifically, LESC employs a low-rank representation of samples in the feature space, and gLESC leverages a tensor multi-rank minimization to further investigate the sample correlations in both the feature space and label space. Benefitting from the sample correlations, the proposed methods can boost the performance of label enhancement. Extensive experiments on 14 benchmark datasets demonstrate the effectiveness and superiority of our methods.
Image classification systems recently made a giant leap with the advancement of deep neural networks. However, these systems require an excessive amount of labeled data to be adequately trained. ...Gathering a correctly annotated dataset is not always feasible due to several factors, such as the expensiveness of the labeling process or difficulty of correctly classifying data, even for the experts. Because of these practical challenges, label noise is a common problem in real-world datasets, and numerous methods to train deep neural networks with label noise are proposed in the literature. Although deep neural networks are known to be relatively robust to label noise, their tendency to overfit data makes them vulnerable to memorizing even random noise. Therefore, it is crucial to consider the existence of label noise and develop counter algorithms to fade away its adverse effects to train deep neural networks efficiently. Even though an extensive survey of machine learning techniques under label noise exists, the literature lacks a comprehensive survey of methodologies centered explicitly around deep learning in the presence of noisy labels. This paper aims to present these algorithms while categorizing them into one of the two subgroups: noise model based and noise model free methods. Algorithms in the first group aim to estimate the noise structure and use this information to avoid the adverse effects of noisy labels. Differently, methods in the second group try to come up with inherently noise robust algorithms by using approaches like robust losses, regularizers or other learning paradigms.
•Label noise is a common problem in real-world datasets.•Noise robust learning techniques are important to achieve state of the art performance.•Many works are proposed in the literature to tackle noisy labels.•Some works aim to estimate underlying noise structure.•Other works try to achieve robustness without explicitly modeling the noise structure.