Memetic Evolution of Classification Ensembles
Creating classification ensembles may be perceived as a regularization technique which aims at improving the generalization capabilities of a classifier. In this paper, we introduce a multi-level memetic algorithm for evolving classification ensembles (they can be either homo- or heterogeneous). First, we evolve the content of such ensembles, and then we optimize the weights (both for the classifiers and for different classes) exploited while voting. The experimental study showed that our memetic algorithm retrieves high-quality heterogeneous ensembles, and can effectively deal with small training sets in multi-class classification.
KeywordsEnsemble classifier Memetic algorithm Classification
This work was supported by the National Science Centre, Poland, under Research Grant No. DEC-2017/25/B/ST6/00474, and JN was partially supported by the Silesian University of Technology under the Grant for young researchers (BKM-556/RAU2/2018).
- 6.Kuncheva, L.I.: Clustering-and-selection model for classifier combination. In: 2000 Proceedings Fourth International Conference on Knowledge-Based Intelligent Engineering Systems and Allied Technologies, vol. 1, pp. 185–188. IEEE (2000)Google Scholar
- 18.Ribalta, P., Marcinkiewicz, M., Nalepa, J.: Segmentation of hyperspectral images using quantized convolutional neural nets. In: Proceedings of IEEE DSD, pp. 260–267 (2018)Google Scholar