1
$\begingroup$

This is a rather calculational question. I am looking for the ideal of relations among a certain set of polynomials. In polynomial ring $k[X_0, X_1, X_2]$, where $k$ is an algebraic closed field, consider the elementary symmetric polynomials $e_1 = X_0 + X_1 + X_2$, $e_2 = X_0 X_1 + X_1 X_2 + X_0 X_2$, and $e_3 = X_0 X_1 X_2$. Then consider the seven polynomials $a = e_3^2$, $b=e_1 e_2 e_3$, $c=e_1^3 e_3$, $d=e_2^3$, $e=e_1^2 e_2^2$, $f = e_1^4 e_2$, and $g=e_1^6$. These are homogeneous polynomials of degree 6. Now I would like to know the ideal $I$ of relations among them, so I am looking for the kernel of the homomorphism of $k$-algebras $k[A, \dots, G] \rightarrow k[a, \dots, g], A \mapsto a, \dots, G \mapsto g$, where $k[A, \dots, G]$ is the polynomial ring in seven variables. So $I$ is homogeneous prime ideal.

I have found that the following eight Polynomials are in this kernel: $B^2 - AE, E^2 -DF, F^2 - EG, BE - CD, BC - AF, BG - CF, DG - EF, C^2 - AG$. Let $J$ be the ideal generated by these polynomials in $k[A, \dots, G]$. Now I also know that the zero sets of $I$ and $J$ in projective space $\mathbb{P}^6(k)$ coincide, e.g. $V(I) = V(J)$, which implies $\sqrt J = I$, by the projective Nullstellensatz of Hilbert. Also, $J$ is properly included in $I$, because for example the ninth polynomial $h = BF - CE$ doesn't lie in $J$, but it lies in $I$.

I was thinking, that $J + (h) = I$ holds. A way to prove this, is to show that $J + (h)$ is a radical ideal. That is where I am stuck, because I am not familiarized enough with techniques testing if an ideal is a radical ideal or not. I have looked for a Groebner basis of $J + (h)$, using reverse lexicographic order, and found that the nine polynomials $a, \dots, h$ already constitute a Groebner basis for this ideal. I don't know if this can help.

So I would like to know whether $J + (h)$ is a radical ideal or not, and if not, how I can find find additional polynomials from $I$ to $a, \dots, h$, so that they altogether generate the ideal of relations $I$ I was first interested in. I would be very thankful for any help!

Many thanks in advance.

$\endgroup$
0
$\begingroup$

In Macaulay2:

R=QQ[a..g]
S=QQ[e1,e2,e3,Degrees=>{1,2,3}]
phi=map(S,R,{e3^2,e1*e2*e3,e1^3*e3,e2^3,e1^2*e2^2,e1^4*e2,e1^6})
I=ker phi
--ideal(f^2-e*g,e*f-d*g,c*f-b*g,e^2-d*f,c*e-b*f,c*d-b*e,c^2-a*g,b*c-a*f,b^2-a*e)
radical I == I --true
J=ideal(f^2-e*g,e*f-d*g,c*f-b*g,e^2-d*f,c*d-b*e,c^2-a*g,b*c-a*f,b^2-a*e)
radical J == I --true
$\endgroup$
0
$\begingroup$

Here is the Magma code:

Q := RationalField();
R<X0,X1,X2> := PolynomialRing(Q,3);
e1 := X0+X1+X2;
e2 := X0*X1+X0*X2+X1*X2;
e3 := X0*X1*X2;
a := e3^2; 
b := e1*e2*e3; 
c := e1^3*e3; 
d := e2^2; 
e := e1^2*e2^2; 
f := e1^4*e2; 
g := e1^6;
L := [a,b,c,d,e,f,g];
S<A,B,C,D,E,F,G> := PolynomialRing(Q,7);
RelationIdeal(L,S); 

The output is:

Ideal of Polynomial ring of rank 7 over Rational Field
Order: Lexicographical
Variables: A, B, C, D, E, F, G
Inhomogeneous, Dimension >0
Basis:
[
    D^3*G - E^3,
    A*E - B^2,
    A*F - B*C,
    A*G - C^2,
    B*F - C*E,
    B*G - C*F,
    E*G - F^2
]

The 7 polynomials listed in the Basis give a full set of generators for the ideal of relations among your quantities $a,\ldots,g$.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.