7.6 KiB
| title | chunk | source | category | tags | date_saved | instance |
|---|---|---|---|---|---|---|
| Belief revision | 5/7 | https://en.wikipedia.org/wiki/Belief_revision | reference | science, encyclopedia | 2026-05-05T14:44:59.576343+00:00 | kb-cron |
Peppas and Williams provided the formal relationship between revision and update. They introduced the Winslett Identity in the Notre Dame Journal of Formal Logic. Dalal the models of
P
{\displaystyle P}
having a minimal Hamming distance to models of
K
{\displaystyle K}
are selected to be the models that result from the change; Satoh similar to Dalal, but distance between two models is defined as the set of literals that are given different values by them; similarity between models is defined as set containment of these differences; Winslett for each model of
K
{\displaystyle K}
, the closest models of
P
{\displaystyle P}
are selected; comparison is done using set containment of the difference; Borgida equal to Winslett's if
K
{\displaystyle K}
and
P
{\displaystyle P}
are inconsistent; otherwise, the result of revision is
K
∧
P
{\displaystyle K\wedge P}
; Forbus similar to Winslett, but the Hamming distance is used. The revision operator defined by Hegner makes
K
{\displaystyle K}
not to affect the value of the variables that are mentioned in
P
{\displaystyle P}
. What results from this operation is a formula
K
′
{\displaystyle K'}
that is consistent with
P
{\displaystyle P}
, and can therefore be conjoined with it. The revision operator by Weber is similar, but the literals that are removed from
K
{\displaystyle K}
are not all literals of
P
{\displaystyle P}
, but only the literals that are evaluated differently by a pair of closest models of
K
{\displaystyle K}
and
P
{\displaystyle P}
according to the Satoh measure of closeness.
== Iterated revision == The AGM postulates are equivalent to a preference ordering (an ordering over models) to be associated to every knowledge base
K
{\displaystyle K}
. However, they do not relate the orderings corresponding to two non-equivalent knowledge bases. In particular, the orderings associated to a knowledge base
K
{\displaystyle K}
and its revised version
K
∗
P
{\displaystyle K*P}
can be completely different. This is a problem for performing a second revision, as the ordering associated with
K
∗
P
{\displaystyle K*P}
is necessary to calculate
K
∗
P
∗
Q
{\displaystyle K*P*Q}
. Establishing a relation between the ordering associated with
K
{\displaystyle K}
and
K
∗
P
{\displaystyle K*P}
has been however recognized not to be the right solution to this problem. Indeed, the preference relation should depend on the previous history of revisions, rather than on the resulting knowledge base only. More generally, a preference relation gives more information about the state of mind of an agent than a simple knowledge base. Indeed, two states of mind might represent the same piece of knowledge
K
{\displaystyle K}
while at the same time being different in the way a new piece of knowledge would be incorporated. For example, two people might have the same idea as to where to go on holiday, but they differ on how they would change this idea if they win a million-dollar lottery. Since the basic condition of the preference ordering is that their minimal models are exactly the models of their associated knowledge base, a knowledge base can be considered implicitly represented by a preference ordering (but not vice versa). Given that a preference ordering allows deriving its associated knowledge base but also allows performing a single step of revision, studies on iterated revision have been concentrated on how a preference ordering should be changed in response of a revision. While single-step revision is about how a knowledge base
K
{\displaystyle K}
has to be changed into a new knowledge base
K
∗
P
{\displaystyle K*P}
, iterated revision is about how a preference ordering (representing both the current knowledge and how much situations believed to be false are considered possible) should be turned into a new preference relation when
P
{\displaystyle P}
is learned. A single step of iterated revision produces a new ordering that allows for further revisions. Two kinds of preference ordering are usually considered: numerical and non-numerical. In the first case, the level of plausibility of a model is representing by a non-negative integer number; the lower the rank, the more plausible the situation corresponding to the model. Non-numerical preference orderings correspond to the preference relations used in the AGM framework: a possibly total ordering over models. The non-numerical preference relation were initially considered unsuitable for iterated revision because of the impossibility of reverting a revision by a number of other revisions, which is instead possible in the numerical case. Darwiche and Pearl formulated the following postulates for iterated revision.
if
α
⊨
μ
{\displaystyle \alpha \models \mu }
then
(
ψ
∗
μ
)
∗
α
≡
ψ
∗
α
{\displaystyle (\psi *\mu )*\alpha \equiv \psi *\alpha }
; if
α
⊨
¬
μ
{\displaystyle \alpha \models \neg \mu }
, then
(
ψ
∗
μ
)
∗
α
≡
ψ
∗
α
{\displaystyle (\psi *\mu )*\alpha \equiv \psi *\alpha }
; if
ψ
∗
α
⊨
μ
{\displaystyle \psi *\alpha \models \mu }
, then
(
ψ
∗
μ
)
∗
α
⊨
μ
{\displaystyle (\psi *\mu )*\alpha \models \mu }
; if
ψ
∗
α
⊭
¬
μ
{\displaystyle \psi *\alpha \not \models \neg \mu }
, then
(
ψ
∗
μ
)
∗
α
⊭
¬
μ
{\displaystyle (\psi *\mu )*\alpha \not \models \neg \mu }
. Specific iterated revision operators have been proposed by Spohn, Boutilier, Williams, Lehmann, and others. Williams also provided a general iterated revision operator.