← All problems

Finiteness of the Tate–Shafarevich group of an elliptic curve

9 years old · traced to

Let KK be a number field with algebraic closure Kˉ\bar K and absolute Galois group GK=Gal⁡(Kˉ/K)G_K=\operatorname{Gal}(\bar K/K). Let EE be an elliptic curve over KK. For each place vv of KK, finite or infinite, let KvK_v be the completion, fix an embedding Kˉ↪Kˉv\bar K\hookrightarrow\bar K_v, and let GKv⊆GKG_{K_v}\subseteq G_K be the corresponding decomposition group. Write

H1(K,E)=H1(GK,E(Kˉ)),H1(Kv,E)=H1(GKv,E(Kˉv))H^1(K,E)=H^1\bigl(G_K,E(\bar K)\bigr),\qquad H^1(K_v,E)=H^1\bigl(G_{K_v},E(\bar K_v)\bigr)

for Galois cohomology with the discrete topology on the coefficients. The Tate–Shafarevich group of E/KE/K is

Ш(E/K)=ker⁡(H1(K,E)⟶∏vH1(Kv,E)),\text{Ш}(E/K)=\ker\Bigl(H^1(K,E)\longrightarrow\prod_{v}H^1(K_v,E)\Bigr),

the kernel of restriction to all places. Its elements correspond to isomorphism classes of torsors (principal homogeneous spaces) for EE over KK that have a KvK_v-point for every place vv.

Conjecture (Shafarevich–Tate). For every number field KK and every elliptic curve E/KE/K, the group Ш(E/K)\text{Ш}(E/K) is finite.

A solution must prove finiteness for all number fields and all elliptic curves. Ш(E/K)\text{Ш}(E/K) is a torsion abelian group, and each Ш(E/K)[m]\text{Ш}(E/K)[m] is finite. So the conjecture is equivalent to the statement that the pp-primary part Ш(E/K)[p∞]\text{Ш}(E/K)[p^\infty] is finite for every prime pp and is zero for all but finitely many pp.

Background

For each m≥1m\ge1 there is an exact sequence

0⟶E(K)/mE(K)⟶Sel⁡(m)(E/K)⟶Ш(E/K)[m]⟶0,0\longrightarrow E(K)/mE(K)\longrightarrow \operatorname{Sel}^{(m)}(E/K)\longrightarrow \text{Ш}(E/K)[m]\longrightarrow0,

where the Selmer group Sel⁡(m)(E/K)\operatorname{Sel}^{(m)}(E/K) is finite and computable. Descent bounds the rank of E(K)E(K) by Selmer ranks, and Ш\text{Ш} measures the failure of this bound to be sharp. If Ш(E/K)[p∞]\text{Ш}(E/K)[p^\infty] is finite for one prime pp, then the rank of E(K)E(K) can be computed by pkp^k-descent for growing kk [2]. Torsors in Ш\text{Ш} are curves of genus 11 that violate the Hasse principle, such as Selmer's cubic 3x3+4y3+5z3=03x^3+4y^3+5z^3=0 over Q\mathbb Q.

Cassels constructed an alternating pairing on Ш(E/K)\text{Ш}(E/K) whose kernel is the maximal divisible subgroup [4]. If Ш(E/K)\text{Ш}(E/K) is finite, its order is therefore a perfect square. Finiteness is also part of the Birch–Swinnerton-Dyer conjecture, whose leading-term formula contains #Ш(E/K)\#\text{Ш}(E/K) [1].

References
  1. J. Tate, "The arithmetic of elliptic curves", Invent. Math. 23 (1974), 179–206. doi:10.1007/BF01389745
  2. J. H. Silverman, The Arithmetic of Elliptic Curves, 2nd ed., Grad. Texts in Math. 106, Springer (2009). doi:10.1007/978-0-387-09494-6
  3. K. Rubin, "Tate–Shafarevich groups and LL-functions of elliptic curves with complex multiplication", Invent. Math. 89 (1987), 527–559. doi:10.1007/BF01388984
  4. J. W. S. Cassels, "Arithmetic on curves of genus 1. IV. Proof of the Hauptvermutung", J. Reine Angew. Math. 211 (1962), 95–112. doi:10.1515/crll.1962.211.95
  5. B. H. Gross and D. B. Zagier, "Heegner points and derivatives of LL-series", Invent. Math. 84 (1986), 225–320. doi:10.1007/BF01388809
  6. V. A. Kolyvagin, "Finiteness of E(Q)E(\mathbb Q) and Ш(E,Q)(E,\mathbb Q) for a subclass of Weil curves", Math. USSR-Izv. 32 (1989), 523–541. doi:10.1070/IM1989v032n03ABEH000779
  7. C. Breuil, B. Conrad, F. Diamond and R. Taylor, "On the modularity of elliptic curves over Q\mathbb Q: wild 3-adic exercises", J. Amer. Math. Soc. 14 (2001), 843–939. doi:10.1090/S0894-0347-01-00370-8
  8. S. Zhang, "Heights of Heegner points on Shimura curves", Ann. of Math. 153 (2001), 27–147. doi:10.2307/2661372
  9. Wikipedia, Tate–Shafarevich group.

Source: Ya-Qing Hu, “Reflecting Numbers of Various Types, I”, arXiv:2207.02509 (2022).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.