Documentation

Atlas.TensorCategories.code.HigherDerivation

structure TensorCategories.HigherDerivationSystem (k : Type w) [Field k] (A : Type w) [Ring A] [Algebra k A] (ε : A →ₐ[k] k) :

A higher derivation system on a k-algebra A augmented by ε : A →ₐ[k] k: a sequence of k-linear functionals χ n : A → k with χ 0 = ε and the Leibniz-type product formula χ_n(ab) = ∑_j (n choose j) χ_j(a) χ_{n-j}(b). Used to prove vanishing of higher derivations on finite-dimensional algebras in characteristic zero.

Instances For
    theorem TensorCategories.HigherDerivationSystem.χ_one_mul {k : Type w} [Field k] {A : Type w} [Ring A] [Algebra k A] {ε : A →ₐ[k] k} (hds : HigherDerivationSystem k A ε) (a b : A) :
    (hds.χ 1) (a * b) = (hds.χ 1) a * ε b + ε a * (hds.χ 1) b

    Leibniz rule at level one: χ_1 is an ε-twisted derivation.

    theorem TensorCategories.HigherDerivationSystem.χ_apply_zero {k : Type w} [Field k] {A : Type w} [Ring A] [Algebra k A] {ε : A →ₐ[k] k} (hds : HigherDerivationSystem k A ε) (a : A) :
    (hds.χ 0) a = ε a

    The zeroth functional χ 0 agrees with the augmentation ε.

    theorem TensorCategories.HigherDerivationSystem.χ_one_one {k : Type w} [Field k] {A : Type w} [Ring A] [Algebra k A] {ε : A →ₐ[k] k} (hds : HigherDerivationSystem k A ε) :
    (hds.χ 1) 1 = 0

    χ 1 vanishes on the unit of A.

    theorem TensorCategories.HigherDerivationSystem.χ_one_algebraMap {k : Type w} [Field k] {A : Type w} [Ring A] [Algebra k A] {ε : A →ₐ[k] k} (hds : HigherDerivationSystem k A ε) (r : k) :
    (hds.χ 1) ((algebraMap k A) r) = 0

    χ 1 vanishes on scalars from k.

    theorem TensorCategories.HigherDerivationSystem.χ_pow_vanish {k : Type w} [Field k] {A : Type w} [Ring A] [Algebra k A] {ε : A →ₐ[k] k} (hds : HigherDerivationSystem k A ε) {a : A} (ha : ε a = 0) (m n : ) (hmn : m < n) :
    (hds.χ m) (a ^ n) = 0

    For a in the augmentation ideal, χ_m (a^n) = 0 whenever m < n.

    theorem TensorCategories.HigherDerivationSystem.χ_pow_diagonal {k : Type w} [Field k] {A : Type w} [Ring A] [Algebra k A] {ε : A →ₐ[k] k} (hds : HigherDerivationSystem k A ε) {a : A} (ha : ε a = 0) (n : ) :
    (hds.χ n) (a ^ n) = n.factorial * (hds.χ 1) a ^ n

    For a in the augmentation ideal, χ_n (a^n) = n! · (χ_1 a)^n.

    theorem TensorCategories.HigherDerivationSystem.coeff_zero_of_sum_pow_eq_zero {k : Type w} [Field k] {A : Type w} [Ring A] [Algebra k A] {ε : A →ₐ[k] k} (hds : HigherDerivationSystem k A ε) [CharZero k] {a : A} (ha : ε a = 0) (hne : (hds.χ 1) a 0) (s : Finset ) (g : k) (hzero : js, g j a ^ j = 0) (i : ) (hi : i s) :
    g i = 0

    In characteristic zero, any nontrivial polynomial relation ∑ g_j • a^j = 0 among the powers of an augmentation-ideal element with χ_1 a ≠ 0 forces all coefficients to vanish.

    theorem TensorCategories.HigherDerivationSystem.powers_linearIndependent {k : Type w} [Field k] {A : Type w} [Ring A] [Algebra k A] {ε : A →ₐ[k] k} (hds : HigherDerivationSystem k A ε) [CharZero k] {a : A} (ha : ε a = 0) (hne : (hds.χ 1) a 0) :
    LinearIndependent k fun (n : ) => a ^ n

    In characteristic zero, the powers of a are k-linearly independent whenever a lies in the augmentation ideal and χ_1 a ≠ 0.

    Main vanishing theorem (used in the proof of Theorem 1.27.4): on a finite-dimensional algebra in characteristic zero, the first higher derivation χ 1 of any system is zero.

    Typeclass packaging the vanishing of χ 1 for every higher derivation system on every finite-dimensional k-algebra, used as a hypothesis in higher-level results.

    Instances