Abstract
Probabilistic belief contraction is an operation that takes a probability distribution P representing a belief state along with an input sentence a representing some information to be removed from this belief state, and outputs a new probability distribution \(P^-_a\). The contracted belief state \(P^-_a\) can be represented as a mixture of two states: the original belief state P, and the resultant state \(P^*_{\lnot a}\) of revising P by \(\lnot a\). Crucial to this mixture is the mixing factor \(\epsilon \) which determines the proportion of P and \(P^*_{\lnot a}\) that are used in this process in a uniform manner. Ideas from information theory such as the principle of minimum cross-entropy have previously been used to motivate the choice of the probabilistic contraction operation. Central to this principle is the Kullback-Leibler (KL) divergence. In an earlier work we had shown that the KL divergence of \(P^-_a\) from P is fully determined by a function whose only argument is the mixing factor \(\epsilon \). In this paper we provide a way of interpreting \(\epsilon \) in terms of a belief ranking mechanism such as epistemic entrenchment that is in consonance with this result. We also provide a much needed justification for why the mixing factor \(\epsilon \) must be used in a uniform fashion by showing that the minimal divergence of \(P^-_{a}\) from P is achieved only when uniformity is respected.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
- 2.
Sentences that have a probability of 1.
- 3.
\(P^+_a\) is simply Bayesian conditioning.
- 4.
One might wonder if the value of \(\epsilon \) is prefixed. We take the view that it is not, and is indeed sensitive to the information a that is being removed.
- 5.
Strictly speaking Gärdenfors epistemic entrenchment is completely relational, and using ordinals in this way is used for convenience only. Our approach may be taken to be closer to Spohn’s degree of beliefs modeled via Ordinal Conditional Functions (Spohn 1988).
- 6.
We assume that \(a \not \equiv k\). The special case when the agent discards all that it believes will need special treatment, and will digress us to the discussion of special forms of belief contraction such as pick contraction and bunch contraction that are not directly relevant to the main contribution of this paper.
- 7.
KL divergence is often defined only when \(Q(w) = 0\) implies \(P(w) = 0\), obviating the need for special conventions such as \(0 / 0 = 0\).
- 8.
This is the same as saying it is not the case that \(Q(\omega ) = \epsilon \cdot P(\omega )\) for all \(\omega \in [k]\).
- 9.
Strictly speaking relations, but we make appropriatete mental adjustments here.
- 10.
We note here in passing that even if a is assigned the same epistemic rank by the two ranking functions, and hence the revised belief sets \(K^*_{\lnot a}\) and \(K^{*'}_{\lnot a}\) are the same, the revised probabilistic states \(P^*_{\lnot a}\) and \(P^{*'}_{\lnot a}\) could be different. Support for this view can be obtained based on the accounts of probabilistic belief revision developed in (Chhogyal et al. 2014).
References
Alchourrón, C.E., Gärdenfors, P., Makinson, D.: On the logic of theory change: partial meet contraction and revision functions. J. Symb. Log. 50(2), 510–530 (1985)
Arló-Costa, H.L., Levi, I.: Contraction: on the decision-theoretical origins of minimal change and entrenchment. Synthese 152(1), 129–154 (2006)
Chhogyal, K., Nayak, A.C., Sattar, A.: On the KL divergence of probability mixtures for belief contraction. In: Proceedings of the 38th German Conference on Artificial Intelligence (KI-2015) (to appear 2015)
Chhogyal, K., Nayak, A., Schwitter, R., Sattar, A.: Probabilistic belief revision via imaging. In: Pham, D.-N., Park, S.-B. (eds.) PRICAI 2014. LNCS, vol. 8862, pp. 694–707. Springer, Heidelberg (2014)
Chhogyal, K., Nayak, A. C., Zhuang, Z., Sattar, A.: Probabilistic belief contraction using argumentation. In: Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence, IJCAI 2015, Buenos Aires, Argentina, July 25–31, 2015, pp. 2854–2860 (2015)
Cover, T.M., Thomas, J.A.: Elements of Information Theory. John Wiley & Sons, New York (1991)
Dubois, D., Lang, J., Prade, H.: Possibilistic logic. In: Gabbay, D.M., Hogger, C.J., Robinson, J.A. (eds.) Handbook of Logic in Artificial Intelligence and Logic Programming, vol. 3, pp. 439–513. Clarendon Press, Oxford (1994)
Gärdenfors, P.: The dynamics of belief: contractions and revisions of probability functions. Topoi 5(1), 29–37 (1986)
Gärdenfors, P.: Knowledge in Flux. Modelling the Dymanics of Epistemic States. MIT Press, Cambridge (1988)
Grove, A.: Two modellings for theory change. J. Philos. Logic 17(2), 157–170 (1988)
Hansson, S.O.: Belief contraction without recovery. Stud. Logica 50(2), 251–260 (1991)
Kern-Isberner, G.: Linking iterated belief change operations to nonmonotonic reasoning. In: Principles of Knowledge Representation and Reasoning: Proceedings of the Eleventh International Conference, KR 2008, pp. 166–176 (2008)
Kullback, S., Leibler, R.A.: On information and sufficiency. Ann. Math. Stat. 22, 79–86 (1951)
Levi, I.: Truth, faillibility and the growth of knowledge in language, logic and method. Boston Stud. Philos. Sci. N.Y., NY 31, 153–174 (1983)
Nayak, A.C.: Iterated belief change based on epistemic entrenchment. Erkenntnis 41, 353–390 (1994)
Nayak, A.C., Goebel, R., Orgun, M.A.: Iterated belief contraction from first principles. In: IJCAI 2007, Proceedings of the 20th International Joint Conference on Artificial Intelligence, Hyderabad, India, January 6–12, 2007, pp. 2568–2573 (2007)
Potyka, N., Beierle, C., Kern-Isberner, G.: Changes of relational probabilistic belief states and their computation under optimum entropy semantics. In: Timm, I.J., Thimm, M. (eds.) KI 2013. LNCS, vol. 8077, pp. 176–187. Springer, Heidelberg (2013)
Ramachandran, R., Ramer, A., Nayak, A.C.: Probabilistic belief contraction. Mind. Mach. 22(4), 325–351 (2012)
Rott, H., Pagnucco, M.: Severe withdrawal (and recovery). J. Philos. Logic 28(5), 501–547 (1999)
Spohn, W.: Ordinal conditional functions: a dynamic theory of epistemic states. In: Harper, W., Skryms, B. (eds.) Causation in Decision, Belief Change, and Statistics, II’, Kluwer, pp. 105–134 (1988)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Chhogyal, K., Nayak, A., Sattar, A. (2015). Probabilistic Belief Contraction: Considerations on Epistemic Entrenchment, Probability Mixtures and KL Divergence. In: Pfahringer, B., Renz, J. (eds) AI 2015: Advances in Artificial Intelligence. AI 2015. Lecture Notes in Computer Science(), vol 9457. Springer, Cham. https://doi.org/10.1007/978-3-319-26350-2_10
Download citation
DOI: https://doi.org/10.1007/978-3-319-26350-2_10
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-26349-6
Online ISBN: 978-3-319-26350-2
eBook Packages: Computer ScienceComputer Science (R0)