Documentation

Mathlib.Algebra.Lie.Loop

Loop Lie algebras and their central extensions #

Given a Lie algebra L, the loop algebra is the Lie algebra of maps from a circle into L. This can mean many different things, e.g., continuous maps, smooth maps, polynomial maps. In this file, we consider the simplest case of polynomial maps, meaning we take a base change with the ring of Laurent polynomials.

Loop Lie algebras admit central extensions attached to invariant inner products on the base Lie algebra. When the base Lie algebra is finite dimensional and simple, the corresponding central extension (with an outer derivation attached) admits an infinite root system with affine Weyl group. These extended Lie algebras are called untwisted affine Kac-Moody Lie algebras.

We implement the basic theory using AddMonoidAlgebra instead of LaurentPolynomial for flexibility. The classical loop algebra is then written loopAlgebra R ℤ L.

Main definitions #

TODO #

Tags #

lie ring, lie algebra, base change, Laurent polynomial

@[reducible, inline]
abbrev LieAlgebra.loopAlgebra (R : Type u_1) (A : Type u_2) (L : Type u_3) [CommRing R] [LieRing L] [LieAlgebra R L] :
Type (max (max u_2 u_1) u_3)

A loop algebra is the base change of a Lie algebra L over R by R[z,z⁻¹]. We make a slightly more general definition which coincides with the Laurent polynomial construction when A = ℤ

Equations
Instances For

    An Lie algebra isomorphism between the Loop algebra (with A = ℤ) and the tensor product with Laurent polynomials.

    Equations
    Instances For
      def LieAlgebra.LoopAlgebra.toFinsupp (R : Type u_1) (A : Type u_2) (L : Type u_3) [CommRing R] [LieRing L] [LieAlgebra R L] :

      A linear isomorphism to finitely supported functions.

      Equations
      Instances For
        @[simp]
        theorem LieAlgebra.LoopAlgebra.toFinsupp_symm_single (R : Type u_1) (A : Type u_2) (L : Type u_3) [CommRing R] [LieRing L] [LieAlgebra R L] (c : A) (z : L) :
        @[simp]
        theorem LieAlgebra.LoopAlgebra.toFinsupp_single_tmul (R : Type u_1) (A : Type u_2) (L : Type u_3) [CommRing R] [LieRing L] [LieAlgebra R L] (c : A) (z : L) :

        The residue pairing on the loop algebra. When A = ℤ and the elements are viewed as Laurent polynomials with coefficients in L, the pairing is interpreted as (f, g) ↦ Res f dg.

        Equations
        • One or more equations did not get rendered due to their size.
        Instances For
          @[simp]
          theorem LieAlgebra.LoopAlgebra.residuePairing_apply_apply (R : Type u_1) (A : Type u_2) (L : Type u_3) [CommRing R] [LieRing L] [LieAlgebra R L] [AddCommGroup A] [DistribSMul A R] [SMulCommClass A R R] (Φ : LinearMap.BilinForm R L) (f g : loopAlgebra R A L) :
          ((residuePairing R A L Φ) f) g = ((toFinsupp R A L) g).sum fun (a : A) (v : L) => a (Φ (((toFinsupp R A L) f) (-a))) v

          A 2-cochain on a loop algebra given by an invariant bilinear form. When A = ℤ, the alternating condition amounts to the fact that Res f df = 0.

          Equations
          • One or more equations did not get rendered due to their size.
          Instances For
            @[simp]
            theorem LieAlgebra.LoopAlgebra.twoCochainOfBilinear_apply_apply (R : Type u_1) (A : Type u_2) (L : Type u_3) [CommRing R] [LieRing L] [LieAlgebra R L] [CommRing A] [IsAddTorsionFree R] [Algebra A R] (Φ : LinearMap.BilinForm R L) ( : Φ.IsSymm) (x y : loopAlgebra R A L) :
            ((twoCochainOfBilinear R A L Φ ) x) y = (TrivialLieModule.equiv R (loopAlgebra R A L) R).symm (((residuePairing R A L Φ) x) y)

            A 2-cocycle on a loop algebra given by an invariant bilinear form.

            Equations
            Instances For
              @[simp]
              theorem LieAlgebra.LoopAlgebra.twoCocycleOfBilinear_coe (R : Type u_1) (A : Type u_2) (L : Type u_3) [CommRing R] [LieRing L] [LieAlgebra R L] [CommRing A] [IsAddTorsionFree R] [Algebra A R] (Φ : LinearMap.BilinForm R L) ( : LinearMap.BilinForm.lieInvariant L Φ) (hΦs : Φ.IsSymm) :
              (twoCocycleOfBilinear R A L Φ hΦs) = twoCochainOfBilinear R A L Φ hΦs