Advertisement

Learning Complex Concepts Using Crowdsourcing: A Bayesian Approach

  • Paolo Viappiani
  • Sandra Zilles
  • Howard J. Hamilton
  • Craig Boutilier
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6992)

Abstract

We develop a Bayesian approach to concept learning for crowdsourcing applications. A probabilistic belief over possible concept definitions is maintained and updated according to (noisy) observations from experts, whose behaviors are modeled using discrete types. We propose recommendation techniques, inference methods, and query selection strategies to assist a user charged with choosing a configuration that satisfies some (partially known) concept. Our model is able to simultaneously learn the concept definition and the types of the experts. We evaluate our model with simulations, showing that our Bayesian strategies are effective even in large concept spaces with many uninformative experts.

Keywords

Bayesian Approach Recommender System Latent Concept Current Belief True Concept 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Angluin, D.: Queries and concept learning. Machine Learning 2, 319–342 (1988)MathSciNetGoogle Scholar
  2. 2.
    Boutilier, C., Regan, K., Viappiani, P.: Online feature elicitation in interactive optimization. In: Proceedings of the Twenty-sixth International Conference on Machine Learning (ICML 2009), Montreal, pp. 73–80 (2009)Google Scholar
  3. 3.
    Boutilier, C., Regan, K., Viappiani, P.: Simultaneous elicitation of preference features and utility. In: Proceedings of the Twenty-fourth AAAI Conference on Artificial Intelligence (AAAI 2010), Atlanta, pp. 1160–1167 (2010)Google Scholar
  4. 4.
    Chen, S., Zhang, J., Chen, G., Zhang, C.: What if the irresponsible teachers are dominating? In: Proceedings of the Twenty-Fourth AAAI Conference on Artificial Intelligence (AAAI 2010), Atlanta, pp. 419–424 (2010)Google Scholar
  5. 5.
    Dai, P., Mausam, Weld, D.S.: Decision-theoretic control of crowd-sourced workflows. In: Proceedings of the Twenty-Fourth AAAI Conference on Artificial Intelligence (AAAI 2010), Atlanta, pp. 1168–1174 (2010)Google Scholar
  6. 6.
    Haussler, D.: Learning conjunctive concepts in structural domains. Machine Learning 4, 7–40 (1989)Google Scholar
  7. 7.
    Heckerman, D., Horvitz, E., Middleton, B.: An approximate nonmyopic computation for value of information. IEEE Transactions on Pattern Analysis and Machine Intelligence 15(3), 292–298 (1993)CrossRefGoogle Scholar
  8. 8.
    Howard, R.: Information value theory. IEEE Transactions on Systems Science and Cybernetics 2(1), 22–26 (1966)CrossRefGoogle Scholar
  9. 9.
    Kearns, M.J., Li, M.: Learning in the presence of malicious errors. SIAM Journal on Computing 22, 807–837 (1993)MathSciNetCrossRefzbMATHGoogle Scholar
  10. 10.
    Mitchell, T.M.: Version spaces: A candidate elimination approach to rule learning. In: Proceedings of the Fifth International Joint Conference on Artificial Intelligence (IJCAI 1977), Cambridge, pp. 305–310 (1977)Google Scholar
  11. 11.
    Shahaf, D., Horvitz, E.: Generalized task markets for human and machine computation. In: Proceedings of the Twenty-Fourth AAAI Conference on Artificial Intelligence (AAAI 2010), Atlanta, pp. 986–993 (2010)Google Scholar
  12. 12.
    Sutton, R.S., Barto, A.G.: Reinforcement Learning: An Introduction. MIT Press, Cambridge (1998)Google Scholar
  13. 13.
    Viappiani, P., Boutilier, C.: Optimal Bayesian recommendation sets and myopically optimal choice query sets. In: Advances in Neural Information Processing Systems (NIPS), Vancouver, vol. 23, pp. 2352–2360 (2010)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Paolo Viappiani
    • 1
  • Sandra Zilles
    • 2
  • Howard J. Hamilton
    • 2
  • Craig Boutilier
    • 3
  1. 1.Department of Computer ScienceAalborg UniversityDenmark
  2. 2.Department of Computer ScienceUniversity of ReginaCanada
  3. 3.Department of Computer ScienceUniversity of TorontoCanada

Personalised recommendations