Let \(S\) be a numerical semigroup. The Hilbert series or generating function associated to \(S\) is \(H_S(x)=\sum_{s\in S}x^s\) (actually it is the Hilbert function of the ring \(K[S]\) with \(K\) a field). See for instance [Mor14].
‣ NumericalSemigroupPolynomial( s, x ) | ( function ) |
s is a numerical semigroups and x a variable (or a value to evaluate in). The output is the polynomial \(1+(x-1)\sum_{s\in \mathbb N\setminus S} x^s\), which equals \((1-x)H_S(x)\).
gap> x:=X(Rationals,"x");; gap> s:=NumericalSemigroup(5,7,9);; gap> NumericalSemigroupPolynomial(s,x); x^14-x^13+x^12-x^11+x^9-x^8+x^7-x^6+x^5-x+1
‣ IsNumericalSemigroupPolynomial( f ) | ( function ) |
f is a polynomial in one variable. The output is true if there exists a numerical semigroup \(S\) such that f equals \((1-x)H_S(x)\), that is, the polynomial associated to \(S\) (false otherwise).
gap> x:=X(Rationals,"x");; gap> s:=NumericalSemigroup(5,6,7,8);; gap> f:=NumericalSemigroupPolynomial(s,x); x^10-x^9+x^5-x+1 gap> IsNumericalSemigroupPolynomial(f); true
‣ NumericalSemigroupFromNumericalSemigroupPolynomial( f ) | ( function ) |
f is a polynomial associated to a numerical semigroup (otherwise yields error). The output is the numerical semigroup \(S\) such that f equals \((1-x)H_S(x)\).
gap> x:=X(Rationals,"x");; gap> s:=NumericalSemigroup(5,6,7,8);; gap> f:=NumericalSemigroupPolynomial(s,x); x^10-x^9+x^5-x+1 gap> NumericalSemigroupFromNumericalSemigroupPolynomial(f)=s; true
‣ HilbertSeriesOfNumericalSemigroup( s, x ) | ( function ) |
s is a numerical semigroup and x a variable (or a value to evaluate in). The output is the series \(\sum_{s\in S} x^s\). The series is given as a rational function.
gap> x:=X(Rationals,"x");; gap> s:=NumericalSemigroup(5,7,9);; gap> HilbertSeriesOfNumericalSemigroup(s,x); (x^14-x^13+x^12-x^11+x^9-x^8+x^7-x^6+x^5-x+1)/(-x+1)
‣ GraeffePolynomial( p ) | ( function ) |
p is a polynomial. Computes the Graeffe polynomial of p. Needed to test if p is a cyclotomic polynomial (see [BD89]).
gap> x:=Indeterminate(Rationals,1);; SetName(x,"x"); gap> GraeffePolynomial(x^2-1); x^2-2*x+1
‣ IsCyclotomicPolynomial( p ) | ( function ) |
p is a polynomial. Detects if p is a cyclotomic polynomial using the procedure given in [BD89].
gap> CyclotomicPolynomial(Rationals,3); x^2+x+1 gap> IsCyclotomicPolynomial(last); true
‣ IsKroneckerPolynomial( p ) | ( function ) |
p is a polynomial. Detects if p is a Kronecker polynomial, that is, a monic polynomial with integer coefficients having all its roots in the unit circumference, or equivalently, a product of cyclotomic polynomials. The current implementation has been done with A. Herrera-Poyatos, following [BD89].
gap> x:=X(Rationals,"x");; gap> s:=NumericalSemigroup(3,5,7);; gap> t:=NumericalSemigroup(4,6,9);; gap> p:=NumericalSemigroupPolynomial(s,x); x^5-x^4+x^3-x+1 gap> q:=NumericalSemigroupPolynomial(t,x); x^12-x^11+x^8-x^7+x^6-x^5+x^4-x+1 gap> IsKroneckerPolynomial(p); false gap> IsKroneckerPolynomial(q); true
‣ IsCyclotomicNumericalSemigroup( s ) | ( function ) |
s is a numerical semigroup. Detects if the polynomial associated to s is a Kronecker polynomial.
gap> l:=CompleteIntersectionNumericalSemigroupsWithFrobeniusNumber(21);; gap> ForAll(l,IsCyclotomicNumericalSemigroup); true
‣ IsSelfReciprocalUnivariatePolynomial( p ) | ( function ) |
p is a univariate polynomial. Detects if p is selfreciprocal. A numerical semigroup is symmetric if and only if it is selfreciprocal, [Mor14]. The current implementation is due to A. Herrera-Poyatos.
gap> l:=IrreducibleNumericalSemigroupsWithFrobeniusNumber(13);; gap> x:=X(Rationals,"x");; gap> ForAll(l, s-> > IsSelfReciprocalUnivariatePolynomial(NumericalSemigroupPolynomial(s,x))); true
Let \(f(x,y)\in \mathbb K[x,y]\), with \(\mathbb K\) an algebraically closed field of characteristic zero. Let \(f(x,y)=y^n+a_1(x)y^{n-1}+\dots+a_n(x)\) be a nonzero polynomial of \(\mathbb K[x][y]\). After possibly a change of variables, we may assume that, that \(\deg_x(a_i(x))\le i-1\) for all \(i\in\{1,\ldots, n\}\). For \(g\in\mathbb K[x,y]\) that is not a multiple of \(f\), define \(\mathrm{int}(f,g)=\dim_\mathbb K \frac{\mathbb K[x,y]}{(f,g)}\). If \(f\) has one place at infinity, then the set \(\{\mathrm{int}(f,g)\mid g\in\mathbb K[x,y]\setminus(f)\}\) is a free numerical semigroup (and thus a complete intersection).
‣ SemigroupOfValuesOfPlaneCurveWithSinglePlaceAtInfinity( f ) | ( function ) |
f is a polynomial in the variables X(Rationals,1) and X(Rationals,2). Computes the semigroup \(\{\mathrm{int}(f,g)\mid g\in\mathbb K[x,y]\setminus(f)\}\), where \(\mathrm{int}(f,g)=\dim_{\mathbb K} (\mathbb K[x,y]/(f,g))\). The algorithm checks if f has one place at infinity. If the extra argument "all" is given, then the output is the \(\delta\)-sequence and approximate roots of f. The method is explained in [AGS16].
gap> x:=Indeterminate(Rationals,1);; SetName(x,"x"); gap> y:=Indeterminate(Rationals,2);; SetName(y,"y"); gap> f:=((y^3-x^2)^2-x*y^2)^4-(y^3-x^2);; gap> SemigroupOfValuesOfPlaneCurveWithSinglePlaceAtInfinity(f,"all"); [ [ 24, 16, 28, 7 ], [ y, y^3-x^2, y^6-2*x^2*y^3+x^4-x*y^2 ] ]
‣ IsDeltaSequence( l ) | ( function ) |
l is a list of positive integers. Assume that l equals \(a_0,a_1,\dots,a_h\). Then l is a \(\delta\)-sequence if \(\gcd(a_0,\ldots, a_h)=1\), \(\langle a_0,\cdots, a_s\rangle\) is free, \(a_kD_k > a_{k+1}D_{k+1}\) and \(a_0> a_1 > D_2 > D_3 > \ldots > D_{h+1}\), where \(D_1=a_0\), \(D_k=\gcd(D_{k-1},a_{k-1})\).
Every \(\delta\)-sequence generates a numerical semigroup that is the semigroup of values of a plane curve with one place at infinity.
gap> IsDeltaSequence([24,16,28,7]); true
‣ DeltaSequencesWithFrobeniusNumber( f ) | ( function ) |
f is a positive integer. Computes the set of all \(\delta\)-sequences generating numerical semigroups with Frobenius number f.
gap> DeltaSequencesWithFrobeniusNumber(21); [ [ 8, 6, 11 ], [ 10, 4, 15 ], [ 12, 8, 6, 11 ], [ 14, 4, 11 ], [ 15, 10, 4 ], [ 23, 2 ] ]
‣ CurveAssociatedToDeltaSequence( l ) | ( function ) |
l is a \(\delta\)-sequence. Computes a curve in the variables X(Rationals,1) and X(Rationals,2) whose semigroup of values is generated by the l.
gap> CurveAssociatedToDeltaSequence([24,16,28,7]); y^24-8*x^2*y^21+28*x^4*y^18-56*x^6*y^15-4*x*y^20+70*x^8*y^12+24*x^3*y^17-56*x^\ 10*y^9-60*x^5*y^14+28*x^12*y^6+80*x^7*y^11+6*x^2*y^16-8*x^14*y^3-60*x^9*y^8-24\ *x^4*y^13+x^16+24*x^11*y^5+36*x^6*y^10-4*x^13*y^2-24*x^8*y^7-4*x^3*y^12+6*x^10\ *y^4+8*x^5*y^9-4*x^7*y^6+x^4*y^8-y^3+x^2 gap> SemigroupOfValuesOfPlaneCurveWithSinglePlaceAtInfinity(last,"all"); [ [ 24, 16, 28, 7 ], [ y, y^3-x^2, y^6-2*x^2*y^3+x^4-x*y^2 ] ]
‣ SemigroupOfValuesOfPlaneCurve( f ) | ( function ) |
f is a polynomial in the variables X(Rationals,1) and X(Rationals,2). The singular package is mandatory. Either by loading it prior to numerical semigroups or by using NumSgpsUseSingular(). If \(f\) is irreducible, computes the semigroup \(\{\mathrm{int}(f,g)\mid g\in\mathbb K[x,y]\setminus(f)\}\), where \(\mathrm{int}(f,g)=\dim_{\mathbb K} (\mathbb K[\![x,y]\!]/(f,g))\). If it has two components, the output is the value semigroup in two variables, and thus a good semigroup. If there are more components, then the output is that of semigroup in the alexpoly singular library.
gap> x:=X(Rationals,"x");; y:=X(Rationals,"y");; gap> f:= y^4-2*x^3*y^2-4*x^5*y+x^6-x^7; -x^7+x^6-4*x^5*y-2*x^3*y^2+y^4 gap> SemigroupOfValuesOfPlaneCurve(f); <Numerical semigroup with 3 generators> gap> MinimalGenerators(last); [ 4, 6, 13 ] gap> f:=(y^4-2*x^3*y^2-4*x^5*y+x^6-x^7)*(y^2-x^3);; gap> SemigroupOfValuesOfPlaneCurve(f); <Good semigroup> gap> MinimalGenerators(last); [ [ 4, 2 ], [ 6, 3 ], [ 13, 15 ], [ 29, 13 ] ]
‣ SemigroupOfValuesOfCurve_Local( arg ) | ( function ) |
The function admits one or two parameters. In any case, the first is a list of polynomials pols. And the second can be the string "basis" or an integer val.
If only one argument is given, the output is the semigroup of all possible orders of \(K[[pols]]\) provided that \(K[[x]]/K[[pols]]\) has finite length. If the second argument "basis" is given, then the output is a (reduced) basis of the algebra \(K[[pols]]\) such that the orders of the basis elements generate minimally the semigroup of orders of \(K[[pols]]\). If an integer val is the second argument, then the output is a polynomial in \(K[[pols]]\) with order val (fail if there is no such polynomial, that is, val is not in the semigroup of values).
The method is explained in [AGSM17].
gap> x:=Indeterminate(Rationals,"x");; gap> SemigroupOfValuesOfCurve_Local([x^4,x^6+x^7,x^13]); <Numerical semigroup with 4 generators> gap> MinimalGeneratingSystem(last); [ 4, 6, 13, 15 ] gap> SemigroupOfValuesOfCurve_Local([x^4,x^6+x^7,x^13], "basis"); [ x^4, x^7+x^6, x^13, x^15 ] gap> SemigroupOfValuesOfCurve_Local([x^4,x^6+x^7,x^13], 20); x^20
‣ SemigroupOfValuesOfCurve_Global( arg ) | ( function ) |
The function admits one or two parameters. In any case, the first is a list of polynomials pols. And the second can be the string "basis" or an integer val.
If only one argument is given, the output is the semigroup of all possible degrees of \(K[pols]\) provided that \(K[x]/K[pols]\) has finite length. If the second argument "basis" is given, then the output is a (reduced) basis of the algebra \(K[pols]\) such that the degrees of the basis elements generate minimally the semigroup of degrees of \(K[pols]\). If an integer val is the second argument, then the output is a polynomial in \(K[pols]\) with degree val (fail if there is no such polynomial, that is, val is not in the semigroup of values).
The method is explained in [AGSM17].
gap> x:=Indeterminate(Rationals,"x");; gap> SemigroupOfValuesOfCurve_Global([x^4,x^6+x^7,x^13]); <Numerical semigroup with 3 generators> gap> MinimalGeneratingSystem(last); [ 4, 7, 13 ] gap> SemigroupOfValuesOfCurve_Global([x^4,x^6+x^7,x^13],"basis"); [ x^4, x^7+x^6, x^13 ] gap> SemigroupOfValuesOfCurve_Global([x^4,x^6+x^7,x^13],12); x^12 gap> SemigroupOfValuesOfCurve_Global([x^4,x^6+x^7,x^13],6); fail
‣ GeneratorsModule_Global( A, M ) | ( function ) |
A and M are lists of polynomials in the same variable. The output is a basis of the ideal \(M K[A]\), that is, a set \(F\) such that \(deg(F)\) generates the ideal \(deg(M K[A])\) of \(deg(K[A])\), where \(deg\) stands for degree. The method is explained in [AAGS17].
gap> t:=Indeterminate(Rationals,"t");; gap> A:=[t^6+t,t^4];; gap> M:=[t^3,t^4];; gap> GeneratorsModule_Global(A,M); [ t^3, t^4, t^5, t^6 ]
‣ GeneratorsKahlerDifferentials( A, M ) | ( function ) |
A is a list of polynomials in the same variable. The output is GeneratorsModule_Global(A,M), with M the set of derivatives of the elements in A.
gap> t:=Indeterminate(Rationals,"t");; gap> GeneratorsKahlerDifferentials([t^3,t^4]); [ t^2, t^3 ]
‣ IsMonomialNumericalSemigroup( S ) | ( property ) |
S is a numerical semigroup. Tests whether S a monomial numerical semigroup.
Let \(R\) a Noetherian ring such that \(K \subseteq R \subseteq K[[t]]\), \(K\) is a field of characteristic zero, the algebraic closure of \(R\) is \(K[[t]]\), and the conductor \((R : K[[t]])\) is not zero. If \(v : K((t))\to {\mathbb Z}\) is the natural valuation for \(K((t))\), then \(v(R)\) is a numerical semigroup.
Let \(S\) be a numerical semigroup minimally generated by \(\{n_1,\ldots,n_e\}\). The semigroup ring associated to \(S\) is \(K[[S]]=K[[t^{n_1},\ldots,t^{n_e}]]\). A ring is called a semigroup ring if it is of the form \(K[[S]]\), for some numerical semigroup \(S\). We say that \(S\) is a monomial numerical semigroup if for any \(R\) as above with \(v(R)=S\), \(R\) is a semigroup ring. See [Mic02] for details.
gap> IsMonomialNumericalSemigroup(NumericalSemigroup(4,6,7)); true gap> IsMonomialNumericalSemigroup(NumericalSemigroup(4,6,11)); false
generated by GAPDoc2HTML