Nullstellensatz - Algebraic Curves - Solved Exam, Exams of Algebra

This is the Solved Exam of Algebraic Curves which includes Nullstellensatz, Radical, Algebraically, Variety, General, Ideal, Characteristic, Non Singular etc. Key important points are: Nullstellensatz, Radical, Algebraically, Variety, General, Ideal, Characteristic, Non Singular, Coordinates, Group Law

Typology: Exams

2012/2013

Uploaded on 02/18/2013

sankait
sankait ๐Ÿ‡ฎ๐Ÿ‡ณ

4.2

(13)

113 documents

1 / 6

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
MA40188
University of Bath
DEPARTMENT OF MATHEMATICAL SCIENCES
EXAMINATION
MA40188: ALGEBRAIC CURVES
May 2009
No calculators may be brought in and used.
Full marks will be given for correct answers to THREE questions.
Only the best three answers will contribute towards the assessment.
MA40188
pf3
pf4
pf5

Partial preview of the text

Download Nullstellensatz - Algebraic Curves - Solved Exam and more Exams Algebra in PDF only on Docsity!

MA

University of Bath

DEPARTMENT OF MATHEMATICAL SCIENCES

EXAMINATION

MA40188: ALGEBRAIC CURVES

May 2009

No calculators may be brought in and used.

Full marks will be given for correct answers to THREE questions. Only the best three answers will contribute towards the assessment.

MA

MA40188 2.

  1. (a) If K is an algebraically closed field and I is an ideal of K[x 1 ,... , xn] such that V (I) = โˆ… in AnK , then 1 โˆˆ I. [3, bookwork] (b)

I = {a โˆˆ R | an^ โˆˆ I for some n โˆˆ N}. It is an ideal because if an^ โˆˆ I and r โˆˆ R then (ra)n^ โˆˆ I and if an, bm^ โˆˆ I then

(a + b)n+m^ =

โˆ‘ (n + m r

arbn+mโˆ’r^ โˆˆ I

since either r โ‰ฅ n or n + m โˆ’ r โ‰ฅ m. [4, unseen but hint given in lectures.] (c) V (I) = {(a 1 ,... , an) โˆˆ An^ | f (a 1 ,... , an) = 0 for all f โˆˆ I}, and I(V ) = {f โˆˆ K[x 1 ,... , xn] | f (a 1 ,... , an) for all a 1 ,... an โˆˆ V }. If n = 1 and I = ใ€ˆx^2 ใ€‰ then V (I) = { 0 } โŠ‚ A^1 but I(V (I)) = ใ€ˆxใ€‰. [3, bookwork] (d) Suppose f โˆˆ A = K[x 1 ,... , xn]. Consider the ring B = A[y] = K[x 1 ,... , xn, y], and the ideal I+^ = IB + (yf โˆ’ 1)B of B. Notice that Q โˆˆ An+1^ is in V (I+) if and only if the point P โˆˆ An^ got by taking the first n coordinates of Q is in V (I) and, in addition, the last coordinate of Q is 1 /f (P ) (in particular f (P ) 6 = 0). The set (f 6 = 0) โŠ‚ V (I) is empty when f = 0 everywhere on V (I), i.e. when f โˆˆ I(V (I)). So suppose f (P ) = 0 for all P โˆˆ V (I): that means that V (I+) = โˆ…. By the Nullstellensatz, that implies that 1 โˆˆ I+, and because I+^ is generated by I and yf โˆ’ 1 we can find polynomials g 0 , g 1 ,... gk โˆˆ B such that g 0 (yf โˆ’ 1) + g 1 f 1 + ยท ยท ยท + gkfk = 1, where f 1 ,... , fk are generators for the ideal I. This equation is an identity, so writing 1/f instead of y we have

โˆ‘^ k

i=

gi

x 1 ,... , xn, 1 /f (x 1 ,... , xn)

fi(x 1 ,... , xn) = 1.

The left-hand side is a rational function with denominator f N^ where N is the maximum of the degrees of the gi in y), so

gi

x 1 ,... , xn, 1 /f (x 1 ,... , xn)

= hi(x 1 ,... , xn)/

f (x 1 ,... , xn)

)N

for some polynomials hi. If we multiply through by f N^ we get

โˆ‘^ k

i=

hi(x 1 ,... , xn, 1)fi(x 1 ,... , xn) = f (x 1 ,... , xn)N

so f โˆˆ

I as claimed. [10, bookwork]

Question 2 continued...

(d) First, P โˆˆ E because b = 19 = โˆ’4 and 1^3 + 9 ร— 1 โˆ’ 4 = 6 = 11^2 (mod 23). By the formula, the x-coordinate of โˆ’ 2 P is

(1^2 โˆ’ 9)^2 โˆ’ 8 ร— 1 ร— (โˆ’4) 24

= 4 mod 23.

So the x-coordinate of 4P is (4^2 โˆ’ 9)^2 โˆ’ 8 ร— 4 ร— (โˆ’4) 4 ร— (4^3 + 9 ร— 4 โˆ’ 4)

49 + 4 ร— 32

49 + 4 ร— 9

Therefore 4P = ยฑP , but if 4P = P then 2P = โˆ’P ; but we have already seen that P and 2P have different x-coordinates, whereas P and โˆ’P have the same x-coordinate. So 4P = โˆ’P , so 5P = 0. [7, unseen]

  1. (a) A rational map ฯ† : V 99K W is given by ฯ† = (f 0 :... : fn) with fi โˆˆ K[x 0 ,... , xn] all homogeneous of the same degree, such that the fi are not all in the homogenous ideal of V and ฯ†(x) โˆˆ W if x โˆˆ V and ฯ†(x) is defined. [3, bookwork] (b) V and W are birationally equivalent if there exist rational maps ฯ† : V 99K W and ฯˆ : W 99K V such that ฯˆ โ—ฆ ฯ† and ฯ† โ—ฆ ฯˆ are the identity where they are defined. [2, bookwork] (c) V is rational if V is birationally equivalent to some Pr. [2, bookwork] (d) P โˆˆ V is singular if dim TP V > dim TQV for some Q โˆˆ V. [2, bookwork] (e) The singular points of the curve C in P^2 given by

f = x^2 (x โˆ’ y)(x + y)z + x^5 + 3y^5 = 0

are found by setting z = 1 and fx = fy = 0 (writing fx for โˆ‚fโˆ‚x ), and similarly for y and z. It is easiest to begin with y = 1. Then f = x^4 z โˆ’ x^2 z + x^5 + 3, so fz = x^4 โˆ’ x^2 and fx = 4x^3 z โˆ’ 2 x + 5x^4. The equation fz = 0 gives x = 0, x = 1 or x = โˆ’1: but none of these satisfy both f = 0 and fx = 0. If y 6 = 1 then y = 0 and on that line the equation is x^4 z + x^5 = 0, so x = 0 or x = โˆ’z, i.e. the points (0 : 0 : 1) and (โˆ’1 : 0 : 1). So we can check these on the z = 1 part, where we have f = x^4 โˆ’ x^2 y^2 + x^5 + 3y^5 , fx = 4x^3 โˆ’ 2 xy^2 + 5x^4 and fy = โˆ’ 2 x^2 y + 15y^4. At the point (โˆ’1 : 0 : 1), fx does not vanish so that is not a singular point, but all three vanish at (0 : 0 : 1) which is thus the only singular point of C. [6, unseen] (f) Projecting from the singular point gives a birational map ฯ€ : C 99K P^1. We may do this on the part z = 1, since the line z = 0 is not contained in C. Then the line of slope t has y = tx and passes through C where x^4 (1 โˆ’ t^2 ) + x^5 (1 + 3t^5 ) = 0, so the unique nonzero point is at x = t (^2) โˆ’ 1 3 t^5 +1 and this gives a birational map^ P

1 99K C.

inverse to ฯ€. [5, unseen]