Naive height of an algebraic number #
The naive height is the largest absolute value of a coefficient of the primitive integer minimal polynomial. It is defined in every characteristic-zero field, without a chosen complex embedding, and is constant on conjugacy classes. The value at a transcendental element is one.
For rational numbers it agrees with the multiplicative Weil height. For general
algebraic numbers it instead measures the size of their defining polynomial;
its bounds control denominators and the norms of real or complex roots, and
norm_aeval_primitiveMinpoly_le bounds the primitive minimal polynomial near its
complex roots.
References #
- J.-H. Evertse, Diophantine Approximation (2019), ch. 3, §3.1, Exercise 3.1, and ch. 6, §6.1.
Naive height is the native coefficient sup norm of the primitive minimal polynomial.
The naive height is the supremum of the absolute values of the coefficients.
Transcendental elements have naive height one.
The naive height bounds each coefficient of the primitive minimal polynomial.
The positive leading coefficient is bounded by the naive height.
Naive height is at least one, including at transcendental elements.
The natural denominator is at most the naive height. For a transcendental element the
natural denominator is the junk value 0.
On rational numbers naive height agrees with the multiplicative Weil height.
Inversion preserves naive height, also at zero and at transcendental elements.
The norm of a real or complex algebraic number is bounded by its naive height plus one.
A nonzero real or complex algebraic number is bounded away from zero by its naive height.
Near a complex algebraic number α of degree d, its primitive minimal polynomial F_α is
small: |F_α(z)| ≤ (d + 1) (2 + |z|) ^ d H(α) |z - α|, where H(α) is the naive height of α.
The bound is only informative when |z - α| ≤ 1, but holds for every z.