Non-Bayesian Learning in the Presence of Byzantine Agents

Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9888)

Abstract

This paper addresses the problem of non-Bayesian learning over multi-agent networks, where agents repeatedly collect partially informative observations about an unknown state of the world, and try to collaboratively learn the true state. We focus on the impact of the Byzantine agents on the performance of consensus-based non-Bayesian learning. Our goal is to design an algorithm for the non-faulty agents to collaboratively learn the true state through local communication.

We propose an update rule wherein each agent updates its local beliefs as (up to normalization) the product of (1) the likelihood of the cumulative private signals and (2) the weighted geometric average of the beliefs of its incoming neighbors and itself (using Byzantine consensus). Under mild assumptions on the underlying network structure and the global identifiability of the network, we show that all the non-faulty agents asymptotically agree on the true state almost surely.

Keywords

Distributed learning Byzantine agreement Fault-tolerance Adversary attacks Security 

References

  1. 1.
    Bajovic, D., Jakovetic, D., Moura, J.M., Xavier, J., Sinopoli, B.: Large deviations performance of consensus+ innovations distributed detection with non-gaussian observations. IEEE Trans. Signal Process. 60(11), 5987–6002 (2012)MathSciNetCrossRefGoogle Scholar
  2. 2.
    Cattivelli, F.S., Sayed, A.H.: Distributed detection over adaptive networks using diffusion adaptation. IEEE Trans. Signal Process. 59(5), 1917–1932 (2011)MathSciNetCrossRefGoogle Scholar
  3. 3.
    Chamberland, J.-F., Veeravalli, V.V.: Decentralized detection in sensor networks. IEEE Trans. Signal Process. 51(2), 407–416 (2003)CrossRefGoogle Scholar
  4. 4.
    Dolev, D., Lynch, N.A., Pinter, S.S., Stark, E.W., Weihl, W.E.: Reaching approximate agreement in the presence of faults. J. ACM 33(3), 499–516 (1986)MathSciNetCrossRefMATHGoogle Scholar
  5. 5.
    Fekete, A.D.: Asymptotically optimal algorithms for approximate agreement. Distrib. Comput. 4(1), 9–29 (1990)MathSciNetCrossRefGoogle Scholar
  6. 6.
    Gale, D., Kariv, S.: Bayesian learning in social networks. Games Econ. Behav. 45(2), 329–346 (2003)MathSciNetCrossRefMATHGoogle Scholar
  7. 7.
    Hajnal, J., Bartlett, M.: Weak ergodicity in non-homogeneous markov chains. In: Mathematical Proceedings of the Cambridge Philosophical Society, vol. 54, pp. 233–246. Cambridge University Press (1958)Google Scholar
  8. 8.
    Jadbabaie, A., Molavi, P., Sandroni, A., Tahbaz-Salehi, A.: Non-bayesian social learning. Games Econ. Behav. 76(1), 210–225 (2012)MathSciNetCrossRefMATHGoogle Scholar
  9. 9.
    Jadbabaie, A., Molavi, P., Tahbaz-Salehi, A.: Information heterogeneity and the speed of learning in social networks. Columbia Business School Research Paper, (13–28) (2013)Google Scholar
  10. 10.
    Jakovetic, D., Moura, J.M., Xavier, J.: Distributed detection over noisy networks: large deviations analysis. IEEE Trans. Signal Process. 60(8), 4306–4320 (2012)MathSciNetCrossRefGoogle Scholar
  11. 11.
    Lalitha, A., Sarwate, A., Javidi, T.: Social learning and distributed hypothesis testing. In: IEEE International Symposium on Information Theory, pp. 551–555. IEEE (2014)Google Scholar
  12. 12.
    Lynch, N.A.: Distributed Algorithms. Morgan Kaufmann, San Francisco (1996)Google Scholar
  13. 13.
    Mendes, H., Herlihy, M.: Multidimensional approximate agreement in Byzantine asynchronous systems. In: Proceedings of the Forty-fifth Annual ACM Symposium on Theory of Computing, STOC 2013, pp. 391–400. ACM, New York (2013)Google Scholar
  14. 14.
    Molavi, P., Tahbaz-Salehi, A., Jadbabaie, A.: Foundations of non-bayesian social learning. Columbia Business School Research Paper (2015)Google Scholar
  15. 15.
    Nedic, A., Olshevsky, A., Uribe, C.A.: Nonasymptotic convergence rates for cooperative learning over time-varying directed graphs. ArXiv e-prints 1410.1977 (2014)Google Scholar
  16. 16.
    Pease, M., Shostak, R., Lamport, L.: Reaching agreement in the presence of faults. J. ACM 27(2), 228–234 (1980)MathSciNetCrossRefMATHGoogle Scholar
  17. 17.
    Perles, M.A., Sigron, M.: A generalization of tverberg’s theorem. ArXiv e-prints 0710.4668 (2007)Google Scholar
  18. 18.
    Rad, K.R., Tahbaz-Salehi, A.: Distributed parameter estimation in networks. In: 49th IEEE Conference on Decision and Control (CDC), pp. 5050–5055. IEEE (2010)Google Scholar
  19. 19.
    Shahrampour, S., Jadbabaie, A.: Exponentially fast parameter estimation in networks using distributed dual averaging. In: 52nd IEEE Conference on Decision and Control, pp. 6196–6201. IEEE (2013)Google Scholar
  20. 20.
    Shahrampour, S., Rakhlin, A., Jadbabaie, A.: Distributed detection: finite-time analysis and impact of network topology (2014)Google Scholar
  21. 21.
    Shahrampour, S., Rakhlin, A., Jadbabaie, A.: Finite-time analysis of the distributed detection problem. In: 2015 53rd Annual Allerton Conference on Communication, Control, and Computing (Allerton), pp. 598–603. IEEE (2015)Google Scholar
  22. 22.
    Su, L., Vaidya, N.: Reaching approximate Byzantine consensus with multi-hop communication. In: Pelc, A., Schwarzmann, A.A. (eds.) SSS 2015. LNCS, vol. 9212, pp. 21–35. Springer, Heidelberg (2015)CrossRefGoogle Scholar
  23. 23.
    Su, L., Vaidya, N.H.: Defending non-Bayesian learning against adversarial attacks. ArXiv e-prints, June 2016Google Scholar
  24. 24.
    Tsitsiklis, J.N.: Decentralized detection by a large number of sensors. Math. Control Signals Syst. 1(2), 167–182 (1988)MathSciNetCrossRefMATHGoogle Scholar
  25. 25.
    Vaidya, N.H.: Iterative Byzantine vector consensus in incomplete graphs. In: Chatterjee, M., Cao, J., Kothapalli, K., Rajsbaum, S. (eds.) ICDCN 2014. LNCS, vol. 8314, pp. 14–28. Springer, Heidelberg (2014)CrossRefGoogle Scholar
  26. 26.
    Vaidya, N.H., Garg, V.K.: Byzantine vector consensus in complete graphs. In: Proceedings of the ACM Symposium on Principles of Distributed Computing, PODC 2013, pp. 65–73. ACM, New York (2013)Google Scholar
  27. 27.
    Vaidya, N.H., Tseng, L., Liang, G.: Iterative approximate Byzantine consensus in arbitrary directed graphs. In: Proceedings of the ACM Symposium on Principles of Distributed Computing, pp. 365–374. ACM (2012)Google Scholar
  28. 28.
    Varshney, P.K.: Distributed bayesian detection: Parallel fusion network. In: Distributed Detection and Data Fusion, pp. 36–118. Springer, New York (1997)Google Scholar
  29. 29.
    Wolfowitz, J.: Products of indecomposable, aperiodic, stochastic matrices. Proc. Am. Math. Soc. 14(5), 733–737 (1963)MathSciNetCrossRefMATHGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2016

Authors and Affiliations

  1. 1.Department of Electrical and Computer EngineeringUniversity of Illinois at Urbana-ChampaignChampaignUSA

Personalised recommendations