Multiple-instance Learning from Triplet Comparison Bags

Shu，Senlin1; Wang，Deng Bao2; Yuan，Suqin3; Wei，Hongxin4; Jiang，Jiuchuan5; Feng，Lei6; Zhang，Min Ling7

2024-02-12

10.1145/3638776

ACM Transactions on Knowledge Discovery from Data   影响因子和分区

1556-4681

1556-472X

Multiple-instance learning (MIL) solves the problem where training instances are grouped in bags, and a binary (positive or negative) label is provided for each bag. Most of the existing MIL studies need fully labeled bags for training an effective classifier, while it could be quite hard to collect such data in many real-world scenarios, due to the high cost of data labeling process. Fortunately, unlike fully labeled data, triplet comparison data can be collected in a more accurate and human-friendly way. Therefore, in this article, we for the first time investigate MIL from only triplet comparison bags, where a triplet (X, X, X ) contains the weak supervision information that bag X is more similar to X than to X. To solve this problem, we propose to train a bag-level classifier by the empirical risk minimization framework and theoretically provide a generalization error bound. We also show that a convex formulation can be obtained only when specific convex binary losses such as the square loss and the double hinge loss are used. Extensive experiments validate that our proposed method significantly outperforms other baselines.

Empirical Risk Minimization Multi-Instance Learning Triplet Comparison

SCI ; EI

英语

其他

National Science Foundation of China[62176055]

Computer Science

Computer Science, Information Systems ; Computer Science, Software Engineering

WOS:001190988100018

ASSOC COMPUTING MACHINERY

2-s2.0-85185815285

Scopus

1.Chongqing University,Chongqing,China
2.Southeast University,Nanjing,China
3.The University of Sydney,Sydney,Australia
4.Southern University of Science and Technology,Shenzhen,China
5.Nanjing University of Finance and Economics,Nanjing,China
6.Nanyang Technological University,Singapore
7.

Shu，Senlin,Wang，Deng Bao,Yuan，Suqin,et al. Multiple-instance Learning from Triplet Comparison Bags[J]. ACM Transactions on Knowledge Discovery from Data,2024,18(4).

Shu，Senlin.,Wang，Deng Bao.,Yuan，Suqin.,Wei，Hongxin.,Jiang，Jiuchuan.,...&Zhang，Min Ling.(2024).Multiple-instance Learning from Triplet Comparison Bags.ACM Transactions on Knowledge Discovery from Data,18(4).

Shu，Senlin,et al."Multiple-instance Learning from Triplet Comparison Bags".ACM Transactions on Knowledge Discovery from Data 18.4(2024).