Exploring Partially Observed Networks with Nonparametric Bandits

  • Kaushalya MadhawaEmail author
  • Tsuyoshi Murata
Conference paper
Part of the Studies in Computational Intelligence book series (SCI, volume 813)


Real-world networks such as social and communication networks are too large to be observed entirely. Such networks are often partially observed such that network size, network topology, and nodes of the original network are unknown. In this paper we formalize the Adaptive Graph Exploring problem. We assume that we are given an incomplete snapshot of a large network and additional nodes can be discovered by querying nodes in the currently observed network. The goal of this problem is to maximize the number of observed nodes within a given query budget. Querying which set of nodes maximizes the size of the observed network? We formulate this problem as an exploration-exploitation problem and propose iKNN-UCB, a novel nonparametric multi-arm bandit (MAB) algorithm for determining which nodes to be queried in an adaptive manner. Using synthetic networks and real-world networks from different domains, we demonstrate that our proposed algorithm discovers up to 40% more nodes compared to existing state-of-the-art algorithms.



This work was supported by JSPS Grant-in-Aid for Scientific Research(B) (Grant Number 17H01785) and JST CREST (Grant Number JPMJCR1687).


  1. 1.
    Ahmed, N.K., Neville, J., Kompella, R.: Network sampling: from static to streaming graphs. ACM Trans. Knowl. Discov. Data (TKDD) 8(2), 7 (2014)Google Scholar
  2. 2.
    Auer, P.: Using confidence bounds for exploitation-exploration trade-offs. J. Mach. Learn. Res. 3(Nov), 397–422 (2002)MathSciNetzbMATHGoogle Scholar
  3. 3.
    Avrachenkov, K., Basu, P., Neglia, G., Ribeiro, B., Towsley, D.: Pay few, influence most: online myopic network covering. In: 2014 IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), pp. 813–818. IEEE (2014)Google Scholar
  4. 4.
    Bilgic, M., Mihalkova, L., Getoor, L.: Active learning for networked data. In: Proceedings of the 27th International Conference on Machine Learning (ICML-10), pp. 79–86 (2010)Google Scholar
  5. 5.
    Dani, V., Hayes, T.P., Kakade, S.M.: Stochastic linear optimization under bandit feedback. In: 21st Annual Conference on Learning Theory, pp. 0–20 (2008)Google Scholar
  6. 6.
    Fortunato, S.: Community detection in graphs. Phys. Rep. 486(3–5), 75–174 (2010)MathSciNetCrossRefGoogle Scholar
  7. 7.
    Guan, M.Y., Jiang, H.: Nonparametric stochastic contextual bandits (2018)Google Scholar
  8. 8.
    Henderson, K., et al.: Rolx: structural role extraction & mining in large graphs. In: Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 1231–1239. ACM (2012)Google Scholar
  9. 9.
    Lancichinetti, A., Fortunato, S., Radicchi, F.: Benchmark graphs for testing community detection algorithms. Phys. Rev. E 78(4), 046,110 (2008)Google Scholar
  10. 10.
    Lee, S.H., Kim, P.J., Jeong, H.: Statistical properties of sampled networks. Phys. Rev. E 73(1), 016,102 (2006)Google Scholar
  11. 11.
    Ma, Y., Huang, T.K., Schneider, J.G.: In: In: UAI, (ed.), pp. 542–551(2015)Google Scholar
  12. 12.
    Mahajan, A., Teneketzis, D.: Multi-armed bandit problems. Foundations and Applications of Sensor Management, pp. 121–151. Springer, Berlin (2008)Google Scholar
  13. 13.
    Pfeiffer III, J.J., Neville, J., Bennett, P.N.: Active exploration in networks: using probabilistic relationships for learning and inference. In: Proceedings of the 23rd ACM International Conference on Information and Knowledge Management, pp. 639–648. ACM (2014)Google Scholar
  14. 14.
    Robbins, H.: Some aspects of the sequential design of experiments. Bull. Am. Math. Soc. 58(5), 527–535 (1952)MathSciNetCrossRefGoogle Scholar
  15. 15.
    Rusmevichientong, P., Tsitsiklis, J.N.: Linearly parameterized bandits. Math. Oper. Res. 35(2), 395–411 (2010)MathSciNetCrossRefGoogle Scholar
  16. 16.
    Soundarajan, S., Eliassi-Rad, T., Gallagher, B., Pinar, A.: Maxreach: reducing network incompleteness through node probes. In: 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), pp. 152–157. IEEE (2016)Google Scholar
  17. 17.
    Soundarajan, S., Eliassi-Rad, T., Gallagher, B., Pinar, A.: \(\varepsilon \)-wgx: adaptive edge probing for enhancing incomplete networks. In: Proceedings of the 2017 ACM on Web Science Conference, pp. 161–170. ACM (2017)Google Scholar
  18. 18.
    Wang, X., Garnett, R., Schneider, J.: Active search on graphs. In: Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 731–738. ACM (2013)Google Scholar
  19. 19.
    Zhao, Y., Wang, G., Yu, P.S., Liu, S., Zhang, S.: Inferring social roles and statuses in social networks. In: Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 695–703. ACM (2013)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Tokyo Institute of TechnologyTokyoJapan

Personalised recommendations