0% found this document useful (0 votes)
81 views8 pages

Problem Set #2 - Elliptic Curves

This document contains Mario Rosete's solutions to 4 problems from a Math 297 problem set assigned by Professor Julius Basilla. Problem 1 shows that linearly dependent complex numbers ω1 and ω2 have an integral combination ω such that ω1 and ω2 are integral multiples of ω. Problem 2 shows that if ω1 and ω2 are linearly dependent over reals but independent over rationals, then there is a non-zero integral linear combination of arbitrarily small modulus. Problem 3 relates the j-invariants of certain elliptic curves to classical constants like ζ(2k). Problem 4 verifies that elliptic curves with j-invariants 0, 1728 have the expected forms, and that the given equation defines an elliptic curve with the given

Uploaded by

Mario Rosete
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
81 views8 pages

Problem Set #2 - Elliptic Curves

This document contains Mario Rosete's solutions to 4 problems from a Math 297 problem set assigned by Professor Julius Basilla. Problem 1 shows that linearly dependent complex numbers ω1 and ω2 have an integral combination ω such that ω1 and ω2 are integral multiples of ω. Problem 2 shows that if ω1 and ω2 are linearly dependent over reals but independent over rationals, then there is a non-zero integral linear combination of arbitrarily small modulus. Problem 3 relates the j-invariants of certain elliptic curves to classical constants like ζ(2k). Problem 4 verifies that elliptic curves with j-invariants 0, 1728 have the expected forms, and that the given equation defines an elliptic curve with the given

Uploaded by

Mario Rosete
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

MATH 297: Problem Set #2

Due on April 2013

Prof. Julius Basilla

Mario T. Rosete III

1
Mario T. Rosete III MATH 297 (Prof. Julius Basilla ): Problem Set #2

Problem 1
Show that if ω1 and ω2 are complex numbers which are linearly dependent over the rationals Q, then there
is an integral combination ω of ω1 and ω2 such that ω1 and ω2 are integral multiples of ω.

If ω1 and ω2 are dependent over the rationals, then for some integers p and q,
p
ω1 = ω2 .
q

We may assume that p and q are relatively prime, so that, for some integers k and n,

kp + nq = 1.

Take ω = nω1 + kω2 . Then

ω1 = (kp + nq)ω1
 
kp
= q ω1 + nω1
q
= q(kω2 + nω1 )
= qω.

Also,

ω2 = (kp + nq)ω1
 
nq
= p ω2 + kω2
p
= p(nω1 + kω2 )
= pω.

Page 2 of 8
Mario T. Rosete III MATH 297 (Prof. Julius Basilla ): Problem Set #2 Problem 1

Problem 2
Show that if ω1 and ω2 are complex numbers linearly dependent over the reals R but independent over the
rationals Q, then there is a non-zero integral linear combination of them of arbitrarily small modulus.

If ω1 and ω2 are linearly dependent over the reals, then for some real number r,

ω1 = rω2 .

We may assume without loss of generality that |r| ≤ 1. On the other hand if ω1 and ω2 are independent
over Q, then whenever p and q 6= 0, pω1 + qω2 6= 0. Now let  > 0. If we take rationals p, q such that

0 < p, q < 2D|ω2|
, where D is the product of the denominators of p and q, then by independence over Q,

pω1 + qω2 6= 0.

Moreover, taking m and k to be the numerators of p and q, respectively, we have,

|mω1 + kω2 | = |D(pω1 + qω2 )|


= |D(prω2 + qω2 )|
= |Dω2 (pr + q)|

 
≤ D|ω2 | r
+
2D|ω2 | 2D|ω2 |


≤ D|ω2 | =
D|ω2 |

Page 3 of 8
Mario T. Rosete III MATH 297 (Prof. Julius Basilla ): Problem Set #2 Problem 2

Problem 3
X 1
Let Gk (ω1 , ω2 ) = .
ω 2k
ω∈Λ\{0}

(a) Show that lim Gk (z, 1) = 2ζ(2k).


Im(z)→∞

(b) Show that Gn (ω1 , ω2 ) is a polynomial with rational coefficients in G2 (ω1 , ω2 ) and G3 (ω1 , ω2 ).

We separate from Gk (z, 1) those z with imaginary part zero and write
∞ ∞
X 1 X 1 X 1
Gk (z, 1) = + +
z 2k n=1 n2k n=1 (−n)2k
Im(z)6=0

As Im(z) → ∞, the first term vanishes. Thus,



X 1
lim Gk (z, 1) = 2 = 2ζ(2k).
Im(z)→∞
n=1
n2k

We look at

℘0 (z)2 = 4℘(z)3 − g2 ℘(z) − g3 .

Differentiating gives us

℘00 (z) = 6℘2 (z) − 0.5g2

If we write

X
℘(z) = z −2 + b(n)z 2n ,
n=1

we may equate like powers of z. For n = 1, 2

b(1) = g2 /20, b(2) = g3 /28.

Moreover for n ≥ 3,
n−1
X
(2n + 3)(n − 2)b(n) = 3 b(k)b(n − 1 − k).
k=1

Page 4 of 8
Mario T. Rosete III MATH 297 (Prof. Julius Basilla ): Problem Set #2 Problem 3

Problem 4
Show that if j 6= 0, 1728, then the equation

36 1
Y 2 + XY = X 3 − x−
j − 1728 j − 1728

defines an elliptic curve with j-invariant j. Show that Y 2 = X 3 − 1 has j-invariant 0 and Y 2 = X 3 − X has
j-invariant 1728.

For the elliptic curve Y 2 = X 3 − 1, we verify that the j-invariant is 0.

17284p3 17284(0)3
j= = = 0.
4p3 + 27q 2 4(0)3 + 27(−1)2

For the elliptic curve Y 2 = X 3 − X, we verify that the j-invariant is 1728.

17284p3 17284(−1)3
j= = = 1728.
4p3 + 27q 2 4(−1)3 + 27(0)2

For j 6= 0, 1728, we look at computing the j-invariant to verify. We have

−36 −1
a1 = 1, a4 = , and a6 =
j − 1728 j − 1728
1728c34
We need to verify that j= 3 .
c4 − c26

We compute b2 , b4 and b6 .

−72 −4
b2 = a21 = 1, b4 = 2a4 = , and b6 = 4a6 = .
j − 1728 j − 1728

We compute c4 and c6 .

1728 j
c4 = b22 − 24b4 = 1 + =
j − 1728 j − 1728

2592 864 1728 j


c6 = −b32 + 36b2 b4 − 216b6 = −1 − + = −1 − =−
j − 1728 j − 1728 j − 1728 j − 1728

Now, we verify the equation


 3
j
1728
j − 1728
j =  3  2
j j
− −
j − 1728 j − 1728

This is equivalent to
 3  2  3
j j j
j −j − = 1728
j − 1728 j − 1728 j − 1728

Problem 4 continued on next page. . . Page 5 of 8


Mario T. Rosete III MATH 297 (Prof. Julius Basilla ): Problem Set #2 Problem 4 (continued)

 3  2
j j
(j − 1728) =j −
j − 1728 j − 1728

2
j3

j
= j −
(j − 1728)2 j − 1728

2
j 3 = j [−j]

j3 = j3

We have verified the equality by showing it to be equivalent to the above.

Page 6 of 8
Mario T. Rosete III MATH 297 (Prof. Julius Basilla ): Problem Set #2 Problem 4

Problem 5
Show that ℘(z) − ℘(nz) has a simple zero at the nonzero points where (n ± 1)(z) = 0 and 2x 6= 0 and a
double zero where (n ± 1)(z) = 2z = 0.

Observe that

℘(z) − ℘(nz) = 0 ⇒ ℘(z) = ℘(nz) ⇒ z = ±nz ⇒ (n ± 1)z = 0.

Moreover,

1 0
(℘(z) − ℘(nz))0 = ℘0 (z) − n℘0 (nz) = (1 − )℘ (z)
n2
 
0 ω1 ω2 ω1 + ω2
But ℘ (z) has zeroes at z ∈ , , =2 (C/Λ) − {0}
2 2 2
Thus, we get a double root if and only if 2z = 0.

Page 7 of 8
Mario T. Rosete III MATH 297 (Prof. Julius Basilla ): Problem Set #2 Problem 5

Problem 6
Show that ℘(3) (z) = 12℘(z)℘0 (z) for all z, where ℘(3) (z) denotes the third derivative of ℘(z).

We use this equation,

℘0 (z)2 = 4℘(z)3 − g2 ℘(z) − g3 .

Getting the derivative of both sides,

2℘0 (z)℘00 (z) = 12℘(z)2 ℘0 (z) − g2 ℘0 (z)


2℘00 (z) = 12℘(z)2 − g2

Differentiating yet again,

2℘(3) (z) = 24℘(z)℘0 (z)


℘(3) (z) = 12℘(z)℘0 (z)

Page 8 of 8

You might also like