Existing methods based on transfer learning leverage auxiliary information to help tail generalization and improve the performance of the tail classes. However, they cannot fully exploit the relationships between auxiliary information and tail classes and bring irrelevant knowledge to the tail classes. To solve this problem, we propose a hierarchical CNN with knowledge complementation, which regards hierarchical relationships as auxiliary information and transfers relevant knowledge to tail classes. First, we integrate semantics and clustering relationships as hierarchical knowledge into the CNN to guide feature learning. Then, we design a complementary strategy to jointly exploit the two types of knowledge, where semantic knowledge acts as a prior dependence and clustering knowledge reduces the negative information caused by excessive semantic dependence (i.e., semantic gaps). In this way, the CNN facilitates the utilization of the two complementary hierarchical relationships and transfers useful knowledge to tail data to improve long-tailed classification accuracy. Experimental results on public benchmarks show that the proposed model outperforms existing methods. In particular, our model improves accuracy by 3.46% compared with the second-best method on the long-tailed tieredImageNet dataset.

Hierarchical classification learning, which organizes data categories into a hierarchical structure, is an effective approach for large-scale classification tasks. The high dimensionality of data feature space, represented in hierarchical class structures, is one of the main research challenges. In addition, the class hierarchy often introduces imbalanced class distributions and causes overfitting. In this paper, we propose a feature selection method based on label distribution learning to address the above challenges. The crux is to alleviate the class imbalance problem and learn a discriminative feature subset for hierarchical classification process. Due to correlation between different class categories in the hierarchical tree structure, sibling categories can provide additional supervisory information for each learning sub tasks, which, in turn, alleviates the problem of under-sampling of minority categories. Therefore, we transform hierarchical labels to a hierarchical label distribution to represent this correlation. After that, a discriminative feature subset is selected recursively, by the common features and label-specific feature constraints, to ensure that downstream classification tasks can achieve the best performance. Experiments and comparisons, using seven well-established feature selection algorithms on six real data sets with different degrees of imbalance, demonstrate the superiority of the proposed method.

The sizes of datasets in terms of the number of samples, features, and classes have dramatically increased in recent years. In particular, there usually exists a hierarchical structure among class labels as hundreds of classes exist in a classification task. We call these tasks hierarchical classification, and hierarchical structures are helpful for dividing a very large task into a collection of relatively small subtasks. Various algorithms have been developed to select informative features for flat classification. However, these algorithms ignore the semantic hyponymy in the directory of hierarchical classes, and select a uniform subset of the features for all classes. In this paper, we propose a new feature selection framework with recursive regularization for hierarchical classification. This framework takes the hierarchical information of the class structure into account. In contrast to flat feature selection, we select different feature subsets for each node in a hierarchical tree structure with recursive regularization. The proposed framework uses parent-child, sibling, and family relationships for hierarchical regularization. By imposing$\ell _{2,1}$-norm regularization to different parts of the hierarchical classes, we can learn a sparse matrix for the feature ranking at each node. Extensive experiments on public datasets demonstrate the effectiveness and efficiency of the proposed algorithms.

In large-scale data classification tasks, it is becoming more and more challenging in finding a true class from a huge amount of candidate categories. Fortunately, a hierarchical structure usually exists in these massive categories. The task of utilizing this structure for effective classification is called hierarchical classification. It usually follows a top-down fashion which predicts a sample from the root node with a coarse-grained category to a leaf node with a fine-grained category. However, misclassification is inevitable if the information is insufficient or large uncertainty exists in the prediction process. In this scenario, we can design a stopping strategy to stop the sample at an internal node with a coarser category, instead of predicting a wrong leaf node. Several studies address the problem by improving performance in terms of hierarchical accuracy and informative prediction. However, all of these researches ignore an important issue: when predicting a sample at the current node, the error is inclined to occur if large uncertainty exists in the next lower level children nodes. In this paper, we integrate this uncertainty into a risk problem: when predicting a sample at a decision node, it will take precipitance risk in predicting the sample to a children node in the next lower level on one hand, and take conservative risk in stopping at the current node on the other. We address the risk problem by designing a Local Bayes Risk Minimization (LBRM) framework, which divides the prediction process into recursively deciding to stop or to go down at each decision node by balancing these two risks in a top-down fashion. Rather than setting a global loss function in the traditional Bayes risk framework, we replace it with different uncertainty in the two risks for each decision node. The uncertainty on the precipitance risk and the conservative risk are measured by information entropy on children nodes and information gain from the current node to children nodes, respectively. We propose a Weighted Tree Induced Error (WTIE) to obtain the predictions of minimum risk with different emphasis on the two risks. Experimental results on various datasets show the effectiveness of the proposed LBRM algorithm.