NLP for CP
Addressing Constraint Programming with Natural Language Processing
Home
Resources
Publications
Correct
predictions are in
blue
. If we detect only a subset of a labelled sentence, we highlight the caught part as
blue
, the missing part
light blue.
False positives
are in
green
and
false negatives
are in
red
.
Problem 035 (Molnar's problem) — Constraint detection
Molnar
-LSB-
11
-RSB-
posed
the
following
problem
.
Given
k
,
construct
two
$
k
\
times
k
$
matrices
,
M1
and
M2
,
of
integers
such
that
the
determinant
of
M1
is
one
,
the
determinant
of
M2
is
$
\
pm
1
$
,
no
entry
in
M1
or
M2
is
$
\
pm
1
$
,
and
each
entry
in
M2
is
the
square
of
the
corresponding
entry
in
M1
.
We
modelled
this
problem
with
matrices
M1
and
M2
of
decision
variables
,
where
each
entry
of
M2
is
constrained
to
be
equal
to
the
square
of
the
corresponding
entry
of
M1
.
The
determinant
of
each
matrix
is
expressed
as
a
single
,
large-arity
constraint
and
bound
to
a
variable
detVar
with
domain
-LCB-
-1
,
1
-RCB-
.
Problem 035 (Molnar's problem) — Detection of the decisions and objects to be modeled
Molnar
-LSB-
11
-RSB-
posed
the
following
problem
.
Given
k
,
construct
two
$
k
\
times
k
$
matrices
,
M1
and
M2
,
of
integers
such
that
the
determinant
of
M1
is
one
,
the
determinant
of
M2
is
$
\
pm
1
$
,
no
entry
in
M1
or
M2
is
$
\
pm
1
$
,
and
each
entry
in
M2
is
the
square
of
the
corresponding
entry
in
M1
.
We
modelled
this
problem
with
matrices
M1
and
M2
of
decision
variables
,
where
each
entry
of
M2
is
constrained
to
be
equal
to
the
square
of
the
corresponding
entry
of
M1
.
The
determinant
of
each
matrix
is
expressed
as
a
single
,
large-arity
constraint
and
bound
to
a
variable
detVar
with
domain
-LCB-
-1
,
1
-RCB-
.
Back to list