Self-Paced Balance Learning for Clinical Skin Disease Recognition
- 12 June 2019
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 31 (8) , 2832-2846
Abstract
Class imbalance is a challenging problem in many classification tasks. It induces biased classification results for minority classes that contain less training samples than others. Most existing approaches aim to remedy the imbalanced number of instances among categories by resampling the majority and minority classes accordingly. However, the imbalanced level of difficulty of recognizing different categories is also crucial, especially for distinguishing samples with many classes. For example, in the task of clinical skin disease recognition, several rare diseases have a small number of training samples, but they are easy to diagnose because of their distinct visual properties. On the other hand, some common skin diseases, e.g., eczema, are hard to recognize due to the lack of special symptoms. To address this problem, we propose a self-paced balance learning (SPBL) algorithm in this paper. Specifically, we introduce a comprehensive metric termed the complexity of image category that is a combination of both sample number and recognition difficulty. First, the complexity is initialized using the model of the first pace, where the pace indicates one iteration in the self-paced learning paradigm. We then assign each class a penalty weight that is larger for more complex categories and smaller for easier ones, after which the curriculum is reconstructed by rearranging the training samples. Consequently, the model can iteratively learn discriminative representations via balancing the complexity in each pace. Experimental results on the SD-198 and SD-260 benchmark data sets demonstrate that the proposed SPBL algorithm performs favorably against the state-of-the-art methods. We also demonstrate the effectiveness of the SPBL algorithm's generalization capacity on various tasks, such as indoor scene image recognition and object classification.Keywords
Funding Information
- NSFC (61876094)
- Natural Science Foundation of Tianjin, China (18JCYBJC15400, 18ZXZNGX00110)
- Open Project Program of the National Laboratory of Pattern Recognition
- Fundamental Research Funds for the Central Universities
This publication has 81 references indexed in Scilit:
- An insight into classification with imbalanced data: Empirical results and current trends on using data intrinsic characteristicsInformation Sciences, 2013
- Analysis of preprocessing vs. cost-sensitive learning for imbalanced classification. Open problems on intrinsic data characteristicsExpert Systems with Applications, 2012
- SMOTE-RSB *: a hybrid preprocessing approach based on oversampling and undersampling for high imbalanced data-sets using SMOTE and rough sets theoryKnowledge and Information Systems, 2011
- Boosting support vector machines for imbalanced data setsKnowledge and Information Systems, 2009
- Do unbalanced data have a negative effect on LDA?Pattern Recognition, 2008
- Training neural network classifiers for medical decision making: The effects of imbalanced datasets on classification performanceNeural Networks, 2008
- Cost-sensitive boosting for classification of imbalanced dataPattern Recognition, 2007
- Class imbalances versus small disjunctsACM SIGKDD Explorations Newsletter, 2004
- Learning from imbalanced data sets with boosting and data generationACM SIGKDD Explorations Newsletter, 2004
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998