Article details

Title: An Agent Based Approach for Avoiding Cascade Failures
Author(s): George Leu   Akira Namatame            

Abstract: In the latest years, our society tends to become more and more dependent on large scale (global) infrastructure networks. Sometimes, attacks on a few important nodes of the network may generate cascade failures which lead to global failure of the system. This paper describes the mechanism for minimizing cascade failures. An agent based method (using Q-Learning) is provided, in order to prevent the failures to propagate over the entire network.

Keywords: Q-learning, router utilization, router freeness.

[1] L. LI, D. ALDERSON, W. WILLINGER, J. DOYLE – A First-Principles Approach to Understanding the Internet’s Router-Level Topology, Proc. of ACM SIGCOMM’04, Portland, OR, 2004
[2] J.A. BOYAN, M.L. LITTMAN – Packet Routing in Dynamically Changing Networks: A Reinforcement Learning Approach, In J.D. Cowan et al. (Eds.), “Advances in NIPS 6”, pp. 671-678, Morgan Kauffman, San Francisco, CA, 1994
[3] D. NEWTH, J. ASH – Evolving Cascading Failure Resilience in Complex Networks, Proc. of the 8th Asia Pacific Symposium on Intelligent and Evolutionary Systems, pp. 125-136, Cairns, Australia, 2004
[4] A.E. MOTTER – Cascade Control and Defense in Complex Networks, Phys. Rev. Lett., Vol. 93, No. 9, Aug. 2004
[5] A.E. MOTTER, Y.C. LAI – Cascade-Based Attacks on Complex Networks, Phys. Rev. E, Vol. 66, No. 6, Dec. 2002
[6] Y.C. LAI, A.E. MOTTER, T. NISHIKAWA – Attacks and Cascades in Complex Networks, In E. Ben-Naim et al. (Eds.), “Lectures Notes in Physics”, Vol. 650, pp. 299-310, Springer, Berlin, 2004