ProbXiv
Explorersign in
Problem archiveProblem record

Statement

The above construction can be generalized to any endomorphism of E defined over K. For the family of quadratic twists of E and p = 2, the elements of Sel2 (Ed /K) can be described more explicitly: they parametrize bi-quadratic extensions over the field K(E[2]) with certain ramification conditions at places v | 2d∆E \infty, and unramified elsewhere. As this characterization suggests, Selp (Ed /K) is a finite set. Even better, its dimension as an Fp -vector space gives an upper bound on the algebraic rank of Ed (K). Therefore,\sqrt{ }it is a natural question to pursue what the distribution of Selp (Ed /K) is as K( d) varies in F (X). The “Poonen-Rains heuristics” [6] predict the desired distribution to be the following: \sqrt{ } \infty j #{K( d) \in F (X) : dimFp Selp (Ed (K)) = j} Y 1 Y p lim = −k \cdot k . X\to\infty #F (X) 1+p p −1 k=0 k=1 over K is defined as   Y Selp (E/K) := Ker H 1 (K, E[p]) \to H 1 (Kv , E)[p] . v place of K From now on, we will abbreviate the above conjectural probability as P R(j) for each non-negative integer j \geq 0. All aforementioned seminal works focus on verifying the Poonen-Rains heuristics. Smith’s work computes the distribution of 2\infty -Selmer groups, whereas Ellenberg, Landesman, and Levy’s works compute the distribution of m-Selmer groups for all integers m coprime to 6 \cdot char(K). [Main Result]: In joint work with Daniel Keliher [1], we pursue the following two generalizations. We note that p is any odd prime, f (x) is a degree 3 polynomial over K, and rad(I) is the radical of the ideal I. (1) Prime cyclic extensions: F (p, X) := L/K : Gal(L/K) \sim = Z/pZ, NmK Q (rad(∆L/K )) \leq X . (2) Superelliptic curves: C(p, X) := Cd /K : dy p = f (x) : NmK Q (rad((d))) \leq X Assuming the extended Riemann hypothesis, we prove the following results. (1) Suppose Gal(K(E[p])/K) \supset SL2 (Fp ). Then j #{L \in F (p, X) : rk(E(L)) − rk(E(K)) \leq (p − 1) \cdot j} X lim inf \geq P R(i). X\to\infty #F (p, X) i=0

Record

Comments

Nothing has been published against this problem yet, and nothing has been said about it.

Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.

Sign in

Solve with an agent

Open the statement in a chat, with the problem and the ground rules already written into the prompt.

This opens a third-party site. Nothing is posted back to ProbXiv and nothing you write there is recorded here — what a model gives you is an attempt, which a person still has to check.