Diversified dictionaries for multi-instance learning. (April 2017)

• Record Type: Journal Article
• Authors: Qiao, Maoying; Liu, Liu; Yu, Jun; Xu, Chang; Tao, Dacheng
• Is Part Of: Pattern recognition. Volume 64(2017:Apr.)
• Page Start: 407
• Page End: 416
• Publication Date: 2017-04
• Subjects: Multi-instance learning -- Diversified learning -- Dictionary learning; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2016.08.026
• Languages: English
• ISSNs: 0031-3203

Abstract

Abstract: Multiple-instance learning (MIL) has been a popular topic in the study of pattern recognition for years due to its usefulness for such tasks as drug activity prediction and image/text classification. In a typical MIL setting, a bag contains a bag-level label and more than one instance/pattern. How to bridge instance-level representations to bag-level labels is a key step to achieve satisfactory classification accuracy results. In this paper, we present a supervised learning method, diversified dictionaries MIL, to address this problem. Our approach, on the one hand, exploits bag-level label information for training class-specific dictionaries. On the other hand, it introduces a diversity regularizer into the class-specific dictionaries to avoid ambiguity between them. To the best of our knowledge, this is the first time that the diversity prior is introduced to solve the MIL problems. Experiments conducted on several benchmark (drug activity and image/text annotation) datasets show that the proposed method compares favorably to state-of-the-art methods. Abstract : Highlights: This paper presents a supervised diversified dictionaries MIL to address the problem of bridging instance-level representations to bag-level labels. The proposed method exploits bag-level label information for training class-specific dictionaries. The proposed method introduces a diversity regulariser into the class-specific dictionaries to avoid ambiguity between them. To the best of our knowledge, this is the first time that the diversity prior is introduced to solve the MIL problems.