Abstract
Many processes experience abrupt changes in their dynamics. This causes problems for some prediction algorithms which assume that the dynamics of the sequence to be predicted are constant, or at least only change slowly over time. In this paper the problem of predicting sequences with sudden changes in dynamics is considered. For a model of multivariate Gaussian data we derive expected generalization error of standard linear Fisher classifier in situation where after unexpected task change, the classification algorithm learns on a mixture of old and new data. We show both analytically and by an experiment that optimal length of learning sequence depends on complexity of the task, input dimensionality, on the power and periodicity of the changes. The proposed solution is to consider a collection of agents, in this case non-linear single layer perceptrons (agents), trained by a memetic like learning algorithm. The most successful agents are voting for predictions. A grouped structure of the agent population assists in obtaining favorable diversity in the agent population. Efficiency of socially organized evolving multi-agent system is demonstrated on an artificial problem.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Oommen, B.J., Masum, H.: Switching models for non-stationary random environments. IEEE Trans. on Systems, Man and Cybernetics 25, 1334–1339 (1995)
Weng, J., McClelland, J., Pentland, A., Sporns, O., Stockman, I., Sur, M., Thelen, E.: Autonomous mental development by robots and animals. Science 291(5504), 599 (2001)
Raudys, S.: Survival of intelligent agents in changing environments. In: Rutkowski, L., Siekmann, J.H., Tadeusiewicz, R., Zadeh, L.A. (eds.) ICAISC 2004. LNCS (LNAI), vol. 3070, pp. 109–117. Springer, Heidelberg (2004)
Kuncheva, L.: Classifier ensembles for changing environments. In: Roli, F., Kittler, J., Windeatt, T. (eds.) MCS 2004. LNCS, vol. 3077, pp. 1–15. Springer, Heidelberg (2004)
Rutkowski, L.: Adaptive probabilistic neural networks for pattern classification in time-varying environment. IEEE Trans. on Neural Networks 15, 811–827 (2004)
Farmer, J.D., Packard, N.H., Perelson, A.S.: The immune system, adaptation, and machine learning. Physica 22D, 187–204 (1986)
Moody, J.: Economic forecasting: challenges and neural network solutions. In: Proc. of the Int. Symposium on Artificial Neural Networks, Hsinchu, Taiwan (1995)
Huang, W., Lai, K.K., Nakamori, Y., Wang, S.: Forecasting foreign exchange rates with artificial neural networks: A review. Int. J. of Information Technology and Decision Making 3(1), 145–165 (2004)
Raudys, S., Zliobaite, I.: Prediction of commodity prices in rapidly changing environments. In: Singh, S., Singh, M., Apte, C., Perner, P. (eds.) ICAPR 2005. LNCS, vol. 3686, pp. 154–163. Springer, Heidelberg (2005)
Figueiredo, R.J.P.: Convergent algorithms for pattern recognition in nonlinearly evolving nonstationary environment. Proceedings of IEEE 56, 188–189 (1968)
Tzypkin, J.Z.: Introduction to the Self-Learning Systems Theory. Nauka, Moscow (1970)
Bartlett, P.L., Ben-David, S., Kulkarni, S.R.: Learning changing concepts by exploiting the structure of change. Machine Learning 41, 153–174 (2000)
Oommen, B.J., Rueda, L.: Stochastic learning-based weak estimation of multinomial random variables and its applications to pattern recognition in non-stationary environments. Pattern Recognition 39, 328–341 (2006)
De Castro, L.N., Timmis, J.: Artificial Immune Systems: A new computational intelligence approach. Springer, London (2001)
Krasnogor, N., Smith, J.: A tutorial for competent memetic algorithms: model, taxonomy, and design issues. IEEE Trans. on Evolutionary Computation 9, 474–488 (2005)
Raudys, S.: Social organization of evolving multiple classifier system functioning in changing environments. Adaptive and Natural Computing Algorithms. In: ICANNGA 2007. LNCS, vol. 4431, 4432, Springer, Heidelberg (2007)
Oommen, B.J., Raghunath, G., Kuipers, B.: Parameter learning from stochastic teachers and stochastic compulsive liars. IEEE Trans. on Systems, Man and Cybernetics 25(B), 820–834 (2006)
Widrow, B., Hoff, M.E.: Adaptive switching circuits. WESCON Convention Record 4, 96–104 (1960)
Haykin, S.: Neural Networks: A comprehensive foundation, 2nd edn. Prentice-Hall, Englewood Cliffs (1999)
Raudys, S.: Evolution and generalization of a single neurone. I. SLP as seven statistical classifiers. Neural Networks 11, 283–296 (1998)
Raudys, S.: Statistical and Neural Classifiers: An integrated approach to design. LNCS. Springer, Heidelberg (2001)
Fukunaga, K.: Introduction to Statistical Pattern Recognition. Academic Press, London (1990)
Raudys, S.: On the amount of a priori information in designing the classification algorithm (in Russian). Engineering Cybernetics N4, 168–174 (1972), see also: http://www.science.mii.lt/mii/raudys/abstracts.html
Wyman, F., Young, D., Turner, D.: A comparison of asymptotic error rate expansions for the sample linear discriminant function. Pattern Recognition 23, 775–783 (1990)
Raudys, S., Young, D.M.: Results in statistical discriminant analysis: a review of the former Soviet Union literature. J. of Multivariate Analysis 89, 1–35 (2004)
Raudys, S.: On the universality of the single-layer perceptron model. In: Rutkowski, L. (ed.) Neural Networks and Soft Computing Physica-Verlag, pp. 79–86. Springer, Heidelberg (2002)
Raudys, S., Amari, S.: Effect of initial values in simple perception. In: IJCNN 1998. Proc. 1998 IEEE World Congress on Computational Intelligence, pp. 1530–1535. IEEE Press, Los Alamitos (1998)
Raudys, S.: An adaptation model for simulation of aging process. Int. J. of Modern Physics C. 13, 1075–1086 (2002)
Raudys, S., Zliobaite, I.: The multi-agent system for prediction of financial time series. In: Rutkowski, L., Tadeusiewicz, R., Zadeh, L.A., Zurada, J.M. (eds.) ICAISC 2006. LNCS (LNAI), vol. 4029, pp. 653–662. Springer, Heidelberg (2006)
Raudys, S., Zliobaitė, I.: Evolving multi-agent system for prediction of slow trends and random fluctuations in the oil prices (in preparation, 2007)
Raudys, S.: Trainable Fusion Rules. I. Large sample size case. Neural Networks 19, 1506–1516 (2006)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2007 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Raudys, S., Mitasiunas, A. (2007). Multi-agent System Approach to React to Sudden Environmental Changes. In: Perner, P. (eds) Machine Learning and Data Mining in Pattern Recognition. MLDM 2007. Lecture Notes in Computer Science(), vol 4571. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-73499-4_61
Download citation
DOI: https://doi.org/10.1007/978-3-540-73499-4_61
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-73498-7
Online ISBN: 978-3-540-73499-4
eBook Packages: Computer ScienceComputer Science (R0)