This Article 
 Bibliographic References 
 Add to: 
Explaining 'Explaining Away'
March 1993 (vol. 15 no. 3)
pp. 287-292

'Explaining away' is a common pattern of reasoning in which the confirmation of one cause of an observed or believed event reduces the need to invoke alternative causes. The opposite of explaining away also an occur, where the confirmation of one cause increases belief in another. A general qualitative probabilistic analysis of intercausal reasoning is provided and the property of the interaction among the causes (product synergy) that determines which form of reasoning is appropriate is identified. Product synergy extends the qualitative probabilistic network (QPN) formalism to support qualitative intercausal inference about the directions of change in probabilistic belief. The intercausal relation also justifies Occam's razor, facilitating pruning in the search for likely diagnoses.

[1] H. Geffner, "On the logic of defaults," inProc. Nat. Conf. Artificial Intell.(St. Paul, MN), 1988, pp. 449-454.
[2] H. Geffner, "Causal theories for nonmonotonic reasoning," inProc. Nat. Conf. Artificial Intell.(Boston, MA), 1990, pp. 524-530.
[3] M. Henrion, "Uncertainty in artificial intelligence: Is probability epistemologically and heuristically adequate?" inExpert Judgment and Expert Systems (NATO ISI Series F)(J. Mumpoweret al., Eds.). Berlin: Springer-Verlag, 1987, pp. 105-130, vol. 35.
[4] M. Henrion, "Some practical issues in constructing belief networks," inUncertainty in Artificial Intelligence 3(L. N. Kanal, T. S. Levitt, and J. F. Lemmer, Eds). Amsterdam: North-Holland, 1989.
[5] W. Xiaoqing and K. Kinoshita, "A testable design of logic circuits under highly observable condition," inProc. Int. Test Conf., Sept. 1990, pp. 955-963.
[6] M. Henrion and M. J. Druzdzel, "Qualitative propagation and scenario-based explanation of probabilistic reasoning," inUncertainty in Artificial Intelligence 6(P. P. Bonissone, M. Henrion, and L. N. Kanal, Eds.). Amsterdam: North-Holland, 1991.
[7] P. R. Milgrom, "Good news and bad news: Representation theorems and applications,"Bell J. Econ., vol. 12, pp. 380-391, 1981.
[8] E. Paek, "A circumscriptive theory for causal and evidential support," inProc. Nat. Conf. Artificial Intell., 1990, pp. 545-549.
[9] J. Pearl, "Embracing causality in default reasoning,"Artificial Intell., vol. 35, pp. 259-271, 1988.
[10] J. Pearl,Probabilistic Reasoning in Intelligent Systems. San Mateo, CA: Morgan Kaufmann, 1988.
[11] J. Pearl, D. Geiger, and T. Verma, "Conditional independence and its representations,"Kybernetika, vol. 25, pp. 33-44, 1989.
[12] R. D. Shachter, "Evidence absorption and propagation through evidence reversals," inProc. Workshop Uncertainly Artificial Intell.(Windsor, Canada), 1989, pp. 303-310.
[13] M. Shwe, B. Middleton, and D. E. Heckerman, "Probabilistic diagnosis using a reformulation of the Internist-1/QMR knowledge base: I. The probabilistic model and inference algorithms,"Methods Inform. Med., vol. 30, pp. 241-255, 1991.
[14] M. P. Wellman,Formulation of Tradeoffs in Planning Under Uncertainty. London: Pitman, 1990.
[15] M. P. Wellman, "Fundamental concepts of qualitative probabilistic networks,"Artificial Intell., vol. 44, pp. 257-303, 1990.
[16] M. P. Wellman and M. Henrion, "Qualitative intercausal relations, or explaining "explaining away," inPrinciples Knowledge Represent. Reasoning: Proc. Sec. Int. Conf., 1991, pp. 535-546.

Index Terms:
causal explanation; general qualitative probabilistic analysis; intercausal reasoning; product synergy; qualitative probabilistic network; inference; probabilistic belief; Occam's razor; belief maintenance; explanation; inference mechanisms; probability; uncertainty handling
M.P. Wellman, M. Henrion, "Explaining 'Explaining Away'," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 15, no. 3, pp. 287-292, March 1993, doi:10.1109/34.204911
Usage of this product signifies your acceptance of the Terms of Use.