TITLE: 
The AGM theory of belief revision


ABSTRACT: 
In this lecture I will introduce the most influential early work
on belief revision (BR), that of Alchourrón, Gärdenfors and Makinson (AGM),
as well as describe a couple of ways in which this foundational work has been
extended in recent years. In the first part I will formalise the problem of BR
and introduce the AGM Postulates for BR. This is AGM's list of "rationality 
requirements", or axioms, which any useful operator for BR is meant to satisfy. 
I will also give an explicit construction for a BR operator in terms of 
subjective binary "plausibility" relation over the set of propositional models, 
and show how this construction exactly captures the essence of the AGM postulates 
in the sense that (i) it satisfies all of them, and (ii) every BR operator 
satisfying the AGM postulates may be represented in terms of such a plausibility 
relation. Then, if time permits, I will go on to introduce a couple of the 
modifications and extensions of AGM's theory which have been proposed, namely 
non-prioritised revision (in which an agent is allowed to reject new incoming 
information) and iterated revision (in which an agent has to deal with a sequence 
of new information, rather than just one as with AGM).



PREREQUISITES:
The basic problem of belief revision will be framed here within the language of basic, 
finite propositional logic. Familiarity with the syntax and model-theoretic semantics 
(propositional truth-valuations) of propositional logic will therefore be useful.


REFERENCES:
For general in-depth treatment of Belief Revision in the propositional setting:
Sven Ove Hansson, “A Textbook of Belief Dynamics”, Kluwer Academic Publishers, 1999.
Thomas Meyer, “Semantic Belief Change”, available online as postscript file from 
http://ksg.meraka.org.za/~tmeyer/thesis/thesis.ps.gz, 1999.

The paper that started it all:
Carlos Alchourrón, Peter Gärdenfors and David Makinson, “On the logic of theory change: 
Partial meet contraction and revision functions”, Journal of Symbolic Logic Vol. 50, 
No.2, pp 510-530, 1985.

For an overview of early work on non-prioritised revision:
Sven Ove Hansson, “A survey of non-prioritized belief revision”, Erkenntnis Vol 50, 
pp 413-427, 1999.

Iterated belief revision:
Adnan Darwiche and Judea Pearl, “On the logic of iterated belief revision”, Artificial 
Intelligence Vol 89, No.1, pp 1-29, 1997.

Richard Booth, Thomas Meyer and Ka-Shu Wong, “A bad day surfing is better than a good 
day working: How to revise a total preorder”, Proceedings of the 10th International 
Conference on Principles of Knowledge Representation and Reasoning (KR 2006), pp 230-238, 2006.