Using the noisy-OR model can be harmful \(\dots\) but it often is not. (English) Zbl 1341.68272
Liu, Weiru (ed.), Symbolic and quantitative approaches to reasoning with uncertainty. 11th European conference, ECSQARU 2011, Belfast, UK, June 29 – July 1, 2011. Proceedings. Berlin: Springer (ISBN 978-3-642-22151-4/pbk). Lecture Notes in Computer Science 6717. Lecture Notes in Artificial Intelligence, 122-133 (2011).
Summary: The noisy-OR model and its generalizations are frequently used for alleviating the burden of probability elicitation upon building Bayesian networks with the help of domain experts. The results from empirical studies consistently suggest that, when compared with a fully expert-quantified network, using the noisy-OR model will just have a minor effect on the performance of a network. In this paper, we address this apparent robustness and investigate its origin. Our results show that ill-considered use of the noisy-OR model can substantially decrease a network’s performance, yet also that the model has broader applicability than it was originally designed for.
For the entire collection see [Zbl 1216.68033].
For the entire collection see [Zbl 1216.68033].
MSC:
| 68T37 | Reasoning under uncertainty in the context of artificial intelligence |
| 60E05 | Probability distributions: general theory |
| 62C10 | Bayesian problems; characterization of Bayes procedures |
References:
| [1] | Druzdzel, M.J., van der Gaag, L.C.: Building probabilistic networks: ”Where do the numbers come from?” Guest Editors Introduction. IEEE Transactions on Knowledge & Data Engineering 12, 481–486 (2000) · doi:10.1109/TKDE.2000.868901 |
| [2] | van der Gaag, L.C., Renooij, S., Witteman, C.L.M., Aleman, B.M.P., Taal, B.G.: Probabilities for a probabilistic network: a case study in oesophageal cancer. Artificial Intelligence in Medicine 25, 123–148 (2002) · doi:10.1016/S0933-3657(02)00012-X |
| [3] | Pearl, J.: Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufmann Publishers, San Francisco (1988) · Zbl 0746.68089 |
| [4] | Bolt, J., van der Gaag, L.C.: An Empirical Study of the Use of the Noisy-Or Model in a Real-Life Bayesian Network. In: Hüllermeier, E., Kruse, R., Hoffmann, F. (eds.) IPMU 2010. CCIS, vol. 80, pp. 11–20. Springer, Heidelberg (2010) · Zbl 1209.68532 · doi:10.1007/978-3-642-14055-6_2 |
| [5] | Oniśko, A., Druzdzel, M.J., Wasyluk, H.: Learning Bayesian network parameters from small data sets: Application of noisy-OR gates. International Journal of Approximate Reasoning 27, 165–182 (2001) · Zbl 0988.68827 · doi:10.1016/S0888-613X(01)00039-1 |
| [6] | Henrion, M.: Some practical issues in constructing belief networks. In: Lemmer, J., Levitt, T., Kanal, L. (eds.) Proceedings of the Third Annual Conference on Uncertainty in Artificial Intelligence (UAI 1987), pp. 161–174. Elsevier, Amsterdam (1987) |
| [7] | Díez, F.J.: Parameter adjustment in Bayes networks. The generalized noisy-OR gate. In: Heckerman, D. (ed.) Proceedings of the 9th Conference on Uncertainty in Artificial Intelligence (UAI 2001), pp. 165–182. Morgan Kaufmann Publishers, San Francisco (2001) |
| [8] | Anand, V., Downs, S.M.: Probabilistic asthma case finding: a noisy-OR reformulation. In: Proceedings AMIA Symposium, pp. 6–10 (2008) |
| [9] | Coupé, V., van der Gaag, L.C.: Properties of sensitivity analysis of Bayesian belief networks. In: Proceedings of the Joint Session of the 6th Prague Symposium of Asymptotic Statistics and the 13th Prague Conference on Information Theory, Statistical Decision Functions and Random Processes, Prague, pp. 81–86 (1999) |
| [10] | Renooij, S., van der Gaag, L.C.: Exploiting evidence-dependent sensitivity bounds. In: Bacchus, F., Jaakkola, T. (eds.) Proceedings of the 21st Conference in Uncertainty in Artificial Intelligence, pp. 485–492. AUAI Press (2005) |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
