Abduction As Belief Revision: A Model of Preferred Explanations

Artificial Intelligence, 2002

We present different constructions for non-prioritized belief revision, that is, belief changes in which the input sentences are not always accepted. First, we present the concept of explanation in a deductive way. Second, we define multiple revision operators with respect to sets of sentences (representing explanations), giving representation theorems. Finally, we relate the formulated operators with argumentative systems and default reasoning frameworks.

We study the relationship between some structural rules for abductive reasoning and preference relations for selection preferred explanations. We prove that explanatory relations having good structural properties can always be defined by orders over formulas.

This paper is an attempt to show that Abduction can be used to model the dynamics of belief change. In this paper, we take the special case of abductions defined over agents belief states as an inference to the best causal explanation. We use the term abduction to refer to both the process of determining potential explanations and the choice of the best. In this paper, we provide a new approach for belief revision based on abduction. This approach is different from the existing approaches in the sense that 1. here, belief revision process is guided by causal explanation 2. we provide a new entrenchment ordering called causal entrenchment ordering as an abduction which provides a better account of non prioritized entrenchment revisions. Since one of the important components of agent's epistemic state is the core belief-which the agent retain all the time. We explain a belief revision situation in which an agent carries forward various operations of belief change based on what she accepts to be the core of her belief systems. We show that causal relevance plays an important role in the ordering of beliefs. The crucial idea is that an agent first seeks causal explanations before accepting the new information and in the next step she integrates new information with the old by means of minimal adjustments.

In general, the best explanation for a given observation makes no promises on how good it is with respect to other alternative explanations. A major de ciency of message-passing schemes for belief revision in Bayesian networks is their inability to generate alternatives beyond the second best. In this paper, we present a general approach based on linear constraint systems that naturally generates alternative explanations in an orderly and highly e cient manner. This approach is then applied to cost-based abduction problems as well as belief revision in Bayesian networks.

Encyclopedia of Philosophy and the Social Sciences, 2013

The relationship between Peircean abduction and the modern notion of Inference to the Best Explanation (IBE) is a matter of dispute. Some philosophers, such as Harman (1965) and Lipton (1991, p. 58, 2004, p. 56), claim that abduction and IBE are virtually the same. Others, however, hold that they are quite different (Hintikka 1998; Minnameier 2004) and there is no link between them (Campos 2009). In this paper, I argue that neither of these views is correct. I show that abduction and IBE have important similarities as well as differences. Moreover, by bringing a historical perspective to the study of the relationship between abduction and IBE-a perspective that is lacking in the literature-I show that their differences can be well understood in terms of two historic developments in the history of philosophy of science: first, Reichenbach's distinction between the context of discovery and the context of justification-and the consequent jettisoning of the context of discovery from philosophy of science-and second, underdetermination of theory by data.

2013

Previous works have discussed how to represent abductive reasoning in a dynamic epistemic logic framework. It has been discussed how to define an abductive problem, how to define an abductive solution and what to do with the best solution, once this one has been selected. The present work discusses two possible ways in which the agent can select 'the best' explanation.

Logic Journal of the IGPL, 2006

We have described the workings of a dynamic abducer, an abductive reasoning agent that continually fuses information from multiple sources to "maintain situation awareness", that is, to maintain a best estimate of the situation based on incoming data. In particular, we have described a special class of problems of entity tracking and re-identification, and described in some detail how a dynamic abducer for this problem is able, under some circumstances, to appropriately "change its mind". and explicitly revise the conclusion of an earlier reasoning step. This ability results from performing "meta-abductive" reasoning in situations where no plausible hypotheses are available for explaining a new observation report. We have also described ASAS-Smart, an implementation of such a dynamic abducer, and illustrated its capabilities by tracing its reasoning on two example cases. Although the present system is in many ways simple, idealized, and domain-specific, we believe that it illustrates principles of reasoning that have broad applicability.

Studies in Applied Philosophy, Epistemology and Rational Ethics, 2012

The paper starts from the claim that every assertion of belief may be analyzed into an assertion about a counterfactual surprise of the believer in front of a dissonant knowledge. The notion of a counterfactual surprise can be usefully related to the well studied notion of Shackle's potential surprise. In §2 the author stresses the difference between explanation and abduction on one side and interpretation on the other, maintaining that interpretation expresses what the interpreter believes to be the best explanation of the interpreted fact. The analysis which is proposed takes for granted the classical distinction between doxa and episteme, i.e. between a subjective and an objective dimension of the epistemic enterprise. However, in §1 belief is defined by making reference to knowledge, so reversing a relation which has been established by a deeply rooted philosophical tradition. §1. The aim of this note is developing some reflections inspired by a basic intuition about belief statements which may be synthesized as follows : 1) Belief statements have an essentially counterfactual character 2) They presuppose the notion of knowledge and are definable in terms of it 3) They have to do with the notion of surprise To be clearer, the analysis of belief which we want to take as a starting point is provided by the following definition: (Def Be) x believes that A = Df x would be surprised if x were to know that not-A

Studies in Fuzziness and Soft Computing, 2002

Expert Systems with Applications, 1998

In recent years, attention has been devoted to abduction, a hypothetical form of non-monotonic reasoning that tries to fit the best 'explanation' to a given observation. In this paper we present a collection of applications of automated abductive reasoning developed in the Center for Artificial Intelligence of the ITESM (Monterrey, Mexico) in the last five years, covering a range from natural language understanding to software re-use.