Differences between revisions 15 and 17 (spanning 2 versions)
Revision 15 as of 2010-06-22 01:42:17
Size: 2855
Comment:
Revision 17 as of 2010-06-22 02:16:44
Size: 3229
Comment:
Deletions are marked like this. Additions are marked like this.
Line 45: Line 45:
p is a prime of good reduction iff $p \nmid$ discriminant. p is a prime of good reduction iff $p$ does not divide the discriminant.
Line 53: Line 53:
Step 1: implement w for $E/\Q$ with N square-free (already done in GP) Step 1: implement w for $E/Q$ with N square-free (already done in GP)
Line 64: Line 64:
$w_p for p \nmid 2,3$ has been done by Rohrlich (somewhat hard to read) see Theorems 2 and 3 which are self-contained $w_p for p$ not dividing $2,3$ has been done by Rohrlich [1] (somewhat hard to read) see Theorems 2 and 3 which are self-contained
Line 66: Line 66:
for $p \mid 3$ this has been done Kobayashi for $p \mid 3$ this has been done Kobayashi [2]
Line 68: Line 68:
for $p \mid 2$ T & V Dokchitser the formulae are really hard - ignore this (!!) for $p \mid 2$ T & V Dokchitser [3] the formulae are really hard - ignore this (!!)
Line 70: Line 70:
Would perhaps be better to do for any p, or even determine w globally (T & V Dokshitser: page 1) Would perhaps be better to do for any p, or even determine w globally (T & V Dokchitser: [4] in the introduction)
Line 72: Line 72:

[1] D. Rohrlich, Galois Theory, elliptic curves, and root numbers,
Compos. Math. 100 (1996), 311--349.

[2] S. Kobayashi, The local root number of elliptic curves with wild
ramification, Math. Ann. 323 (2002), 609--623; available online

[3] http://arxiv.org/abs/math/0612054

[4] http://arxiv.org/abs/0906.1815

Tim Dokchitser (Cambridge University): Complex L-functions and the Birch and Swinnerton-Dyer conjecture

Structure of the course

  • Quick review of Elliptic curves over Q and the Mordell-Weil theorem
  • Elliptic curves over finite fields, heuristics for their distribution and the naive version of BSD
  • L-functions of elliptic curves and the BSD-conjecture
  • Root numbers and how to compute them
  • Parity predictions, Goldfeld's conjecture and ranks of elliptic curves over number fields

Prerequisites

Some familiarity with basic algebraic number theory (number fields, primes), and having seen elliptic curves

Background reading

J. H. Silverman, "The arithmetic of elliptic curves", Chapters 3, 7 and 8.

Sage Reference Manual on elliptic curves: http://sagemath.org/doc/reference/plane_curves.html, up to `Isogenies'.

Computational projects

There will be many small problems and larger assignments to play with, illustrating all the concepts and conjectures from the course.

A. Root Numbers over K for elliptic curves (implement)

  • People: Armin, Charlie, Hatice, Christ, Lola, Robert Miller, Thilina, M. Tip, Robert Bradshaw

B. #III(E/K)_{an} function (L-functions, connection to Wuthrich)

  • People: Berinder, M. Tip, Adam, Robert Miller, Robert Bradshaw, Chris Wuthrich

C. Parity Predictions

  • People: Arijit, Anil, Adam

Computing root numbers project: notes

References: 1. Silverman I Chapter VII (does not mention root numbers but gives background information to compute them) - reduction types of elliptic curves etc; plus Section of Silverman I Appendix C on Tate's algorithm.

Elliptic Curves over Q

root number w = \prod_p w_p * w_\infty

p is a prime of good reduction iff p does not divide the discriminant. If E has good reduction at p then w_p = +1.

p \mid \mid N means p is a prime of multiplicative reduction If E has split multiplicative reduction then w_p = -1 If E has non-split multiplicative reduction then w_p = +1

Step 1: implement w for E/Q with N square-free (already done in GP)

If p^2\mid N then p is a prime of additive reduction for E and w_p is more complicated. There are formulae to compute them, they rely on Tate's algorithm.

Elliptic Curves over general number fields

Root number classification

w = \prod_p w_p \prod_{v\div \infty} (-1)

Additive reduction w_p for p not dividing 2,3 has been done by Rohrlich [1] (somewhat hard to read) see Theorems 2 and 3 which are self-contained

for p \mid 3 this has been done Kobayashi [2]

for p \mid 2 T & V Dokchitser [3] the formulae are really hard - ignore this (!!)

Would perhaps be better to do for any p, or even determine w globally (T & V Dokchitser: [4] in the introduction) There is a decision to be made as to which methods to use.

[1] D. Rohrlich, Galois Theory, elliptic curves, and root numbers, Compos. Math. 100 (1996), 311--349.

[2] S. Kobayashi, The local root number of elliptic curves with wild ramification, Math. Ann. 323 (2002), 609--623; available online

[3] http://arxiv.org/abs/math/0612054

[4] http://arxiv.org/abs/0906.1815

days22/dokchitser (last edited 2010-08-10 11:30:48 by Armin Straub)