Step 37 / 64

University

Galois representations

Encode number-preserving symmetries as transformations of vectors.

Start with a concrete example

Imagine recording each symmetry of an object as instructions for moving arrows. Doing two symmetries in succession must match doing the two recorded arrow moves. That rule of correspondence matters, not just assigning a label.

A small check: try the idea

Think with the concrete example above. You can try again.

Must recording two successive symmetries agree with doing their two recorded arrow moves?

Read the answer and reasoning

Yes; composition must be respected. A representation records actions consistently with their composition. Merely naming each symmetry does not supply that consistency.

Build the idea in three steps

  1. Start with arithmetic symmetries: rearrangements of an extension of the rational numbers that preserve arithmetic and fix rational numbers.
  2. Curve points have an addition operation. The points killed by multiplication by p form the p-torsion; at a prime this has linear structure over residues.
  3. The arithmetic symmetries act on these points. Recording their action as linear transformations is the representation used in this proof.

A small example

“Mod-three representation” concerns an action on three-torsion data. It does not mean checking three sample points.

Lean definition · original excerpt

FLT · aa2d8b3 · Def_FLTPrelim_GaloisRep.lean

noncomputable instance instSMulTorsionBy (n : ℕ) :    SMul (K ≃ₐ[S] K) (Submodule.torsionBy ℤ (W'⁄K).Point n) :=  ⟨fun σ P => ⟨σ • (P : (W'⁄K).Point), smul_mem_torsionBy σ P.property⟩⟩

How to read this Lean

Submodule.torsionBy selects points killed by n. σ is an arithmetic symmetry and • denotes its action on a point.

Where this idea is used

6. The scissors incident9. Two doors to the concert10. A three-note audition11. Borrow a better passport12. The workshops agree13. Pip’s impossible quilt14. The shrinking reservation
I’m lost · Help

Mika’s help desk

Galois representations

One thing at a time

Read the three explanation steps first. The Lean source is there for a closer look afterwards; you do not have to read it first.

In simpler words

Encode number-preserving symmetries as transformations of vectors.

Earlier ideas are listed under “Start with these ideas” on this page.

A small clue

“Mod-three representation” concerns an action on three-torsion data. It does not mean checking three sample points.

Close this panel to return to the exact place you were reading. These are prepared hints for this page.

Start with the first small story →