Ask Your Question
1

Computations in a Quotient Ring

asked 2014-12-16 06:16:23 +0100

Lauren gravatar image

updated 2014-12-16 16:41:18 +0100

slelievre gravatar image

I'm trying to do some computations in a quotient ring in Sage, and I'm having some trouble. For example:

Working with the ring:

R.<x,y,z,w,u,z1,z2,z3,z4,z5> = PolynomialRing(QQ,10)
S.<a,b,c,d,e,m1,m2,m3,m4,m5> = R.quo((x^2,y^2+x*y, z^2+x*z + y*z, w^2 - w*x+w*y, u^2 + u*x+ u*z + u*w))

I want to compute (a*(m3+m4+m5) + b*(m1+m2+m3+m4) + c*(m1+m3) + d*(m1+m2) + e*m1)^5

where I'm thinking about m1, m2, m3, m4, and m5 as arbitrary coefficients. When I type this in it returns

4/19*e^5*m1^5 + 15/19*e^5*m1^4*m2 + 10/19*e^5*m1^3*m2^2 +
10/19*e^5*m1^4*m3 + 40/19*e^5*m1^3*m2*m3 + 15/19*e^5*m1^2*m2^2*m3 -
15/19*e^5*m1*m2^2*m3^2 - 5/19*e^5*m1^2*m3^3 - 10/19*e^5*m1*m2*m3^3 +
5/19*e^5*m1^4*m4 - 15/19*e^5*m1^2*m2^2*m4 - 30/19*e^5*m1*m2^2*m3*m4 -
15/19*e^5*m1^2*m3^2*m4 - 30/19*e^5*m1*m2*m3^2*m4 - 10/19*e^5*m1^3*m4^2 -
15/19*e^5*m1^2*m2*m4^2 - 15/19*e^5*m1^2*m3*m4^2 -
30/19*e^5*m1*m2*m3*m4^2 - 5/19*e^5*m1^4*m5 - 20/19*e^5*m1^3*m2*m5 -
15/19*e^5*m1^2*m2^2*m5 - 20/19*e^5*m1^3*m3*m5 - 60/19*e^5*m1^2*m2*m3*m5
- 30/19*e^5*m1*m2^2*m3*m5 - 15/19*e^5*m1^2*m3^2*m5 -
30/19*e^5*m1*m2*m3^2*m5 - 20/19*e^5*m1^3*m4*m5 - 30/19*e^5*m1^2*m2*m4*m5
- 30/19*e^5*m1^2*m3*m4*m5 - 60/19*e^5*m1*m2*m3*m4*m5

However, this is also equivalent to (some expression of mi's)*a*b*c*d*e.

I want it in this form, because for the problem I'm working on I need this coefficient in front of a*b*c*d*e. But I'm not sure how to ask Sage to convert it to this form? For example, "solve" doesn't seem to work in a quotient ring.

(I'm sorry if this is a silly question. I'm new to Sage!)

edit retag flag offensive close merge delete

Comments

To display lines of code, either indend them four spaces, or select them and click the "code" button (the one with "010 101"). To display fragments of code in a text paragraph, use backticks.

slelievre gravatar imageslelievre ( 2014-12-16 16:39:38 +0100 )edit

1 Answer

Sort by ยป oldest newest most voted
0

answered 2014-12-16 17:00:25 +0100

slelievre gravatar image

updated 2014-12-17 19:47:46 +0100

[Edited 2014-12-17]

Running your code:

sage: R.<x,y,z,w,u,z1,z2,z3,z4,z5> = PolynomialRing(QQ,10)
sage: pp = (x^2, y^2+x*y, z^2+x*z + y*z, w^2 - w*x+w*y, u^2 + u*x+ u*z + u*w)
sage: S.<a,b,c,d,e,m1,m2,m3,m4,m5> = R.quo(pp)

sage: q = (a*(m3+m4+m5) + b*(m1+m2+m3+m4) + c*(m1+m3) + d*(m1+m2) + e*m1)^5
sage: q # outputs a polynomial whose monomials are all (monomial in m1 to m5) * e^5
...

To get that polynomial as a polynomial in a, b, c, d, e, whose coefficients are polynomials in m1, m2, m3, m4, m5, one way is to create another polynomial ring with that kind of structure, and the same variable names; then you can pass the string representation of your polynomial to that ring.

sage: U = PolynomialRing(QQ,['m1', 'm2', 'm3', 'm4', 'm5'])
sage: V = PolynomialRing(U, ['a', 'b', 'c', 'd', 'e'])

sage: qq = V(str(q))
sage: qq
(4/19*m1^5 + ... - 60/19*m1*m2*m3*m4*m5)*e^5

But if you know the result is of the form "(some polynomial in m1 to m5) times a*b*c*d*e", you don't need the auxiliary polynomial ring. You can just do

sage: f = a*b*c*d*e
sage: f
-1/38*e^5

sage: m = q / f
sage: m
-8*m1^5 - 30*m1^4*m2 - 20*m1^3*m2^2 - 20*m1^4*m3 - 80*m1^3*m2*m3 - 30*m1^2*m2^2*m3 + 30*m1*m2^2*m3^2 + 10*m1^2*m3^3 + 20*m1*m2*m3^3 - 10*m1^4*m4 + 30*m1^2*m2^2*m4 + 60*m1*m2^2*m3*m4 + 30*m1^2*m3^2*m4 + 60*m1*m2*m3^2*m4 + 20*m1^3*m4^2 + 30*m1^2*m2*m4^2 + 30*m1^2*m3*m4^2 + 60*m1*m2*m3*m4^2 + 10*m1^4*m5 + 40*m1^3*m2*m5 + 30*m1^2*m2^2*m5 + 40*m1^3*m3*m5 + 120*m1^2*m2*m3*m5 + 60*m1*m2^2*m3*m5 + 30*m1^2*m3^2*m5 + 60*m1*m2*m3^2*m5 + 40*m1^3*m4*m5 + 60*m1^2*m2*m4*m5 + 60*m1^2*m3*m4*m5 + 120*m1*m2*m3*m4*m5

and that's the coefficient you were looking for.

edit flag offensive delete link more

Comments

But e^5 equals -38*a*b*c*d*e in S, so the expression can be written as a polynomial in the mi times a*b*c*d*e

Francis Clarke gravatar imageFrancis Clarke ( 2014-12-17 14:55:42 +0100 )edit

Sorry, my answer didn't correctly reflect what I wanted to express. I edited it and hopefully it is clearer now.

slelievre gravatar imageslelievre ( 2014-12-17 19:48:59 +0100 )edit

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Question Tools

Stats

Asked: 2014-12-16 06:16:23 +0100

Seen: 453 times

Last updated: Dec 17 '14