DIVACE: Diverse and Accurate Ensemble Learning Algorithm
In order for a neural network ensemble to generalise properly, two factors are considered vital. One is the diversity and the other is the accuracy of the networks that comprise the ensemble. There exists a tradeoff as to what should be the optimal measures of diversity and accuracy. The aim of this paper is to address this issue. We propose the DIVACE algorithm which tries to produce an ensemble as it searches for the optimum point on the diversity-accuracy curve. The DIVACE algorithm formulates the ensemble learning problem as a multi-objective problem explicitly.
KeywordsMutual Information Pareto Front Ensemble Member Member Network Neural Computation
Unable to display preview. Download preview PDF.
- 1.Abbass, H.A.: A memetic pareto evolutionary approach to artificial neural networks. In: Proceedings of the 14th Australian Joint Conference on Artificial Intelligence, Berlin, pp. 1–12. Springer, Heidelberg (2000)Google Scholar
- 2.Abbass, H.A.: Pareto neuro-evolution: Constructing ensemble of neural networks using multi-objective optimization. In: The IEEE 2003 Conference on Evolutionary Computation, vol. 3, pp. 2074–2080. IEEE Press, Los Alamitos (2003)Google Scholar
- 5.Brown, G., Wyatt, J., Harris, R., Yao, X.: Diversity creation methods: A survey and categorisation. Journal of Information Fusion (2004) (to appear)Google Scholar
- 11.Storn, R., Price, K.: Differential evolution - a simple and efficient adaptive scheme for global optimization over continuous spaces. Technical Report TR-95-012, International Computer Science Institute, Berkeley, USA (1995)Google Scholar
- 12.Yao, X., Liu, Y.: Evolving neural network ensembles by minimization of mutual information. International Journal of Hybrid Intelligent Systems 1(1) (January 2004)Google Scholar