Positioning a coefficient-height witness as the lexicographic leading coefficient by a bounded integral translation of the affine chart.
theorem
CCDN.lex_leadingCoeff_eq_pure_first
{R : Type u_1}
[CommRing R]
{n d : ℕ}
(f : MvPolynomial (Fin (n + 1)) R)
(hd : f.totalDegree ≤ d)
(hc : f.coeff (Finsupp.single 0 d) ≠ 0)
:
If f has total degree at most d and a nonzero coefficient at x₀ ^ d, then its
lexicographic leading coefficient is that coefficient.
theorem
CCDN.homogeneousTranslation_pure_first
{R : Type u_1}
[CommRing R]
{n d : ℕ}
(a : Fin n → R)
(f : MvPolynomial (Fin (n + 1)) R)
(hf : f.IsHomogeneous d)
:
For f homogeneous of degree d, the coefficient of x₀ ^ d in the translate
homogeneousTranslation R a f is the value f (1, a).
theorem
CCDN.exists_bounded_leading_coefficient_translation
{d : ℕ}
(f : MvPolynomial (Fin 3) ℤ)
(hf : f.IsHomogeneous d)
(hf0 : f ≠ 0)
:
∃ (a : Fin 2 → ℤ),
(∀ (j : Fin 2), 0 ≤ a j ∧ a j ≤ ↑d) ∧ MonomialOrder.lex.leadingCoeff ((homogeneousTranslation ℤ a) f) ≠ 0 ∧ CoefficientHeightLE f
((↑d + 1) ^ (2 * (d + 1)) * |↑(MonomialOrder.lex.leadingCoeff ((homogeneousTranslation ℤ a) f))|)
Every original coefficient is controlled by the leading coefficient obtained by a translation with parameters in 0,...,d.