Groebner
RationalUnivariateRepresentation
compute a rational univariate representation
Calling Sequence
Parameters
Description
Examples
References
RationalUnivariateRepresentation(J, v, opts)
J
-
a list or set of polynomials or a PolynomialIdeal
v
(optional) new variable
opts
optional arguments of the form keyword=value
The RationalUnivariateRepresentation command computes a rational univariate representation (or RUR) for a zero-dimensional ideal J. Zero-dimensional systems have a finite number of complex solutions, and an RUR defines a bijection between those solutions and the roots of a univariate polynomial. The advantage of using this representation is that in the worst case the coefficients are an order of magnitude smaller than those of a lexicographic Groebner basis.
The default output is a sequence consisting of an equation f(v)=0 and a set of substitutions x[i] = u[i](v)/d(v) for each variable x[i]. f(v) is a univariate polynomial defining a common algebraic extension, and the solutions of the system are expressed as rational functions in the new variable v with common denominator d(v). If the v is not specified then the global variable _Z is used by default.
The optional argument output controls the form of the result. output=polynomials returns the RUR in a format that is more suitable for programming. In this case, the command returns a sequence consisting of f(v), d(v), and a list of x[i] = u[i]. Alternatively, output=factored factors the univariate polynomial f(v) and splits the RUR into a union of multiple reduced RURs in each irreducible component of f(v). The output is returned as a sequence of two-element lists each containing f[j](v) and a list of x[i] = rem(u[i], f[j](v))/rem(d(v), f[j](v)) . Note that the list of factors f[j](v) are not necessarily unique within the output; instead, their multiplicity is preserved. Each factor f[j](v) will also be monic.
RationalUnivariateRepresentation does not currently support algebraic extensions (specified by RootOfs or radicals), parameters, or characteristics other than zero.
with⁡Groebner:
F≔5⁢x3−330⁢x⁢y+17,3⁢x2⁢y−20⁢y2+x−2
IsZeroDimensional⁡F
true
GroebnerBasis⁡F,plex⁡x,y
15842000⁢y6+1228200⁢y4−75993⁢y3−33600⁢y2−1770⁢y+285,133500534000⁢y5+2386755720000⁢y4+35538821400⁢y3+211467699989⁢y2+1260279815⁢x−5026814580⁢y−2748131560
RationalUnivariateRepresentation⁡F,v
445⁢v6+12233⁢v3−21780⁢v2−578=0,x=−122330⁢v3+290400⁢v2+115608900⁢v5+122330⁢v2−145200⁢v,y=−1395⁢v4+4400⁢v3+6477⁢v−74808900⁢v5+122330⁢v2−145200⁢v
f,d,N≔RationalUnivariateRepresentation⁡F,v,output=polynomials
f,d,N≔445⁢v6+12233⁢v3−21780⁢v2−578,8900⁢v5+122330⁢v2−145200⁢v,y=−1395⁢v4+4400⁢v3+6477⁢v−7480,x=−122330⁢v3+290400⁢v2+11560
factor⁡f
445⁢v6+12233⁢v3−21780⁢v2−578
with⁡PolynomialIdeals:
J≔F
J≔5⁢x3−330⁢x⁢y+17,3⁢x2⁢y−20⁢y2+x−2
IsPrime⁡J
An example where the univariate polynomial factors:
F≔x2+y2−25,x−72+y−72−25
v2−7⁢v+12=0,x=−24+7⁢v−7+2⁢v,y=−25+7⁢v−7+2⁢v
RationalUnivariateRepresentation⁡F,v,output=factored
v−4,y=3,x=4,v−3,y=4,x=3
A similar system with a single solution of multiplicity two:
F≔x2+y2−25,x−62+y−82−25
v2−6⁢v+9=0,x=v,y=4
v−3,y=4,x=3,v−3,y=4,x=3
Rouillier, F. "Solving zero-dimensional systems through the rational univariate representation." Journal of Applicable Algebra in Engineering, Communication, and Computing, Vol. 9, No. 5 (1999): 433-461.
See Also
Basis
FGLM
IsPrime
IsZeroDimensional
Download Help Document