Multiple Instance Learning with Radial Basis Function Neural Networks
This paper investigates the application of radial basis function neural networks (RBFNs) for solving the problem of multiple instance learning (MIL). As a particular form of the traditional supervised learning paradigm, MIL deals with the classification of patterns grouped into bags. Labels of bags are known but not those of individual patterns. To solve the MIL problem, a neural solution based on RBFNs is proposed. A classical application of RBFNs and bag unit-based variants are discussed. The evaluation, conducted on two benchmark data sets, showed that the proposed bag unit-based variant performs very well.
Unable to display preview. Download preview PDF.
- 1.Andrews, S., Hofmann, T., Tsochantaridis, I.: Multiple Instance Learning with Generalized Support Vector Machines. In: Proc. of 14th Nat. Conf. on A.I, pp. 943–944 (2002)Google Scholar
- 2.Auer, P.: On learning from Multi-instance Examples: Empirical Evaluation of a Theoretical Approach. In: Proc. of 14th Int. Conf. on Machine Learning, pp. 21–29 (1997)Google Scholar
- 3.Chiu, S.: Fuzzy Model Identification Based on Cluster Estimation. Journal of Intelligent and Fuzzy Systems 2(3) (1994)Google Scholar
- 5.Maron, O., Ratan, A.: Multiple-Instance Learning for Natural Scene Classification. In: Proc. of 15th Int. Conf. on Machine Learning, pp. 341–349 (1998)Google Scholar
- 7.Ramon, J., De Raedt, L.: Multi-instance Neural Networks. In: Proc. of the Int. Conf. on Machine Learning Workshop on Attribute-Value and Relational Learning (2000)Google Scholar
- 8.Ray, S., Page, O.: Multiple Instance Regression. In: Proc. of 18th Int. Conf. on Machine Learning, pp. 425–432 (2001)Google Scholar
- 9.Wang, J., Zucker, J.: Solving Multiple-Instance Problem: A Lazy Learning Approach. In: Proc. of 17th Int. Conf. on Machine Learning, pp. 1119–1125 (2000)Google Scholar