Information Technology Reference
In-Depth Information
iff
Ψ
is. Indeed, replacing each variable binding
x
→
s
1
,...,s
n
in a unifier of
Φ
with
x
Ψ
→
seq
(
s
1
,...,s
n
) yields a unifier of
Ψ
,andviceversa.
We can consider
Ψ
as an elementary unification problem in the combined
theory
E
1
∪
E
2
,where
E
1
is a flat equational theory over
{
seq
}
and
V
Ind
,and
E
2
is a free equational theory over
V
Ind
.
E
1
-unification problems are,
in fact, word equations, while
E
2
-unification is Robinson unification. Using the
Baader-Schulz combination method [2], we can prove the following theorem:
F
ix
(
Ψ
)and
Theorem 5.
F
-unifiability of
Ψ
is decidable.
Hence, unifiability of general syntactic sequence unification problem is decid-
able.
As for the unification type, in Example 8 we have seen that
mcu
∅
(
Γ
)is
infinite for
Γ
=
?
∅
. It implies that the syntactic sequence uni-
fication is at least infinitary. To show that it is not nullary, by Proposition 1,
it is sucient to prove existence of an almost minimal set of unifiers for every
syntactic sequence unification problem. We do it in the standard way, by proving
that for any
Γ
, every strictly decreasing chain
σ
1
X,Q
∅
{
f
(
a, x
)
≈
f
(
x, a
)
}
σ
2
X,Q
∅
···
of substitu-
tions in
U
∅
(
Γ
) is finite, where
X
=
V
ar
(
Γ
)and
Q
=
SF
un
(
Γ
). Hence, syntactic
sequence unification is infinitary.
4
Relation with Order-Sorted Higher-Order Unification
Syntactic sequence unification can be considered as a special case of order-sorted
higher-order
E
-unification. Here we show the corresponding encoding in the
framework described in [9]. We consider simply typed
λ
-calculus with the types
i
and
o
. The set of base sorts consists of
ind
,
seq
,
seqc
,
o
such that the type of
o
is
o
and the type of the other sorts is
i
. We will treat individual and sequence
variables as first order variables, sequence functions as second order variables and
define a context Γ such that Γ(
x
)=
ind
for all
x
∈V
Ind
,Γ(
x
)=
seq
for all
x
∈
V
Seq
,Γ(
f
)=
seq
→
seqc
for each
f
∈F
Seq
,andΓ(
f
)=
ind
→···→
ind
Flex
→
n
times
Fix
seqc
for each
f
r
(
f
)=
n
. Individual function symbols are treated
as constants. We assign to each
f
∈F
Seq
with
A
Flex
Ind
∈F
a functional sort
seq
→
ind
and
Fix
to each
f
∈F
Ind
with
A
r
(
f
)=
n
a functional sort
ind
→···→
ind
→
ind
.
n
times
We assume equality constants
≈
s
for every sort
s
.Inaddition,wehavetwo
function symbols: binary
of the sort
seq
→
seq
→
seq
and a constant [] of
the sort
seq
. Sorts are partially ordered as [
ind
≤
seqc
]and[
seqc
≤
seq
]. The
equational theory is an AU-theory, asserting associativity of
with [] as left
and right unit. We consider unification problems for terms of the sort
ind
where
terms are in
βη
-normal form containing no bound variables, and terms whose
head is
are flattened. For a given unification problem
Γ
in this theory, we
are looking for unifiers that obey the following restrictions: If a unifier
σ
binds a
second order variable
f
of the sort
seq
→
seqc
,then
fσ
=
λx.
g
1
(
x
)
,...,g
m
(
x
)
and if
σ
binds a second order variable
f
of the sort
ind
→···→
ind
→
seqc
,
n
times
Search WWH ::
Custom Search