Conjugate of a character:
DefinitionOddOrder_GroupTheory_RepresentationTheory_CharacterConjugatefeit-thompsonfinite-groupsgroup-theoryishida-odd-order
For a finite-dimensional complex representation of a finite group , the character at is the complex conjugate of its value at . The proof avoids eigenvalues: it uses the group-averaged Gram matrix , which is positive definite, to show and then uses cyclicity of the trace.
The bundle consists of lemmas only.
The bundle also proves 1 supporting lemma used by later statements.
Definition code
/-
Copyright (c) 2026 Yawara Ishida. All rights reserved.
Released under Apache 2.0 license as described in the file LICENSE.
Authors: Yawara Ishida
-/
-- From github.com/yawara/odd-order @ 20a6fccbb63c (Apache-2.0; LICENSE at that commit):
-- OddOrder/GroupTheory/RepresentationTheory/CharacterConjugate.lean
-- Generated for Prove2Me by deleting declarations; the remaining commands are verbatim except:
-- `private` dropped where another file uses the declaration, and a solution's main theorem
-- renamed `_root_.solution` (platform convention).
import Mathlib.Analysis.Complex.Basic
import Mathlib.Analysis.RCLike.Basic
import Mathlib.RepresentationTheory.Character
import Mathlib.LinearAlgebra.Matrix.PosDef
import Mathlib.LinearAlgebra.Matrix.NonsingularInverse
namespace OddOrder.RepresentationTheory
open Matrix
open scoped ComplexOrder
variable {G V : Type*} [Group G] [Finite G]
[AddCommGroup V] [Module ℂ V] [FiniteDimensional ℂ V]
/-- For a finite-dimensional complex representation of a finite group,
`χ(g⁻¹) = conj (χ(g))`. -/
theorem character_inv (ρ : Representation ℂ G V) (g : G) :
ρ.character g⁻¹ = star (ρ.character g) := by
classical
have : Fintype G := Fintype.ofFinite G
let b := Module.Free.chooseBasis ℂ V
let Mat : G → Matrix (Module.Free.ChooseBasisIndex ℂ V) (Module.Free.ChooseBasisIndex ℂ V) ℂ :=
fun h => LinearMap.toMatrix b b (ρ h)
have hmul : ∀ h₁ h₂ : G, Mat (h₁ * h₂) = Mat h₁ * Mat h₂ := by
intro h₁ h₂
change LinearMap.toMatrix b b (ρ (h₁ * h₂)) = _
rw [map_mul, LinearMap.toMatrix_mul]
have hone : Mat 1 = 1 := by
change LinearMap.toMatrix b b (ρ 1) = 1
rw [map_one, LinearMap.toMatrix_one]
have hchar : ∀ h : G, ρ.character h = (Mat h).trace := by
intro h
change LinearMap.trace ℂ V (ρ h) = _
rw [LinearMap.trace_eq_matrix_trace ℂ b]
-- every `Mat h` is invertible
have hMatUnit : ∀ h : G, IsUnit (Mat h) := fun h =>
IsUnit.of_mul_eq_one (b := Mat h⁻¹) (by rw [← hmul h h⁻¹, mul_inv_cancel, hone])
set M := Mat g with hMdef
-- `Mat g⁻¹` is the two-sided inverse of `M`.
have hLinv : Mat g⁻¹ * M = 1 := by rw [hMdef, ← hmul g⁻¹ g, inv_mul_cancel, hone]
have hRinv : M * Mat g⁻¹ = 1 := by rw [hMdef, ← hmul g g⁻¹, mul_inv_cancel, hone]
have hMdet : IsUnit M.det :=
IsUnit.of_mul_eq_one (b := (Mat g⁻¹).det)
(by rw [← Matrix.det_mul, hRinv, Matrix.det_one])
have hMinv : M⁻¹ = Mat g⁻¹ := Matrix.inv_eq_left_inv hLinv
-- the group-averaged Gram matrix is positive definite, hence invertible
set S : Matrix (Module.Free.ChooseBasisIndex ℂ V) (Module.Free.ChooseBasisIndex ℂ V) ℂ :=
∑ h : G, (Mat h)ᴴ * Mat h with hSdef
have hSpd : S.PosDef := by
rw [hSdef]
refine Matrix.posDef_sum Finset.univ_nonempty (fun h _ => ?_)
exact Matrix.PosDef.conjTranspose_mul_self (Mat h)
(Matrix.mulVec_injective_of_isUnit (hMatUnit h))
have hSdet : IsUnit S.det := (Matrix.isUnit_iff_isUnit_det S).mp hSpd.isUnit
-- invariance `Mᴴ S M = S` from reindexing the sum by `h ↦ h · g`
have hterm : ∀ h : G, Mᴴ * ((Mat h)ᴴ * Mat h) * M = (Mat (h * g))ᴴ * Mat (h * g) := by
intro h
rw [hmul h g, Matrix.conjTranspose_mul, hMdef]
simp only [Matrix.mul_assoc]
have hSinv : Mᴴ * S * M = S := by
rw [hSdef, Finset.mul_sum, Finset.sum_mul, Finset.sum_congr rfl (fun h _ => hterm h)]
exact Fintype.sum_equiv (Equiv.mulRight g) _ _ (fun x => by rw [Equiv.coe_mulRight])
-- `M⁻¹ = S⁻¹ Mᴴ S`.
have hMinv2 : M⁻¹ = S⁻¹ * Mᴴ * S := by
apply Matrix.inv_eq_left_inv
rw [Matrix.mul_assoc (S⁻¹ * Mᴴ) S M, Matrix.mul_assoc S⁻¹ Mᴴ (S * M),
← Matrix.mul_assoc Mᴴ S M, hSinv, Matrix.nonsing_inv_mul S hSdet]
-- conclude
rw [hchar g⁻¹, hchar g, ← hMinv, hMinv2, Matrix.trace_mul_cycle,
Matrix.mul_nonsing_inv S hSdet, Matrix.one_mul, Matrix.trace_conjTranspose]
end OddOrder.RepresentationTheory
Source
Yawara Ishida, Lean 4 formalization of the Feit–Thompson (odd order) theorem, github.com/yawara/odd-order @ 20a6fccbb63c (Apache-2.0), https://github.com/yawara/odd-order/blob/20a6fccbb63c0b138acda9f7fafd798c42f9441b/OddOrder/GroupTheory/RepresentationTheory/CharacterConjugate.lean