cs.SCAug 28, 2026

Generic Characteristic-Zero Equivalence Between Derivative Bézout Inversion and Multipoint Evaluation

Authors: Zijian Zeng

Organizations: Institute of Computer Science and Digital Innovation, UCSI University, Kuala Lumpur, 56000, MALAYSIA

Abstract

Let a1,…,ama_1,\ldots,a_m be distinct elements of a field KK, and let Z(X)=∏i=1m(X−ai)Z(X)=\prod_{i=1}^m (X-a_i). We study the arithmetic complexity of computing the unique normalized Bezout pair s,ts,t satisfying sZ+tZ′=1sZ+tZ'=1, with °t<m°t<m and °s<m−1°s<m-1. The classical product-tree approach requires O(MK(m)log⁡m)O(M_K(m)\log m) field operations, where MK(m)M_K(m) denotes the cost of multiplying degree-<m<m polynomials over KK. Thus, even when MK(m)=O(mlog⁡m)M_K(m)=O(m\log m), the resulting bound is O(mlog⁡2m)O(m\log^2 m) rather than O(mlog⁡m)O(m\log m). Over an infinite field of characteristic zero, we prove that, in the generic rational straight-line-program model, computing all coefficients of the canonical Bezout pair is equivalent, up to an additive O(MK(m))O(M_K(m)) cost, to arbitrary-node multipoint polynomial evaluation and to interpolation. The main ingredient is an explicit differential reconstruction that recovers ZZ from (s,t)(s,t) in O(MK(m))O(M_K(m)) arithmetic operations on a nonempty Zariski-open subset. Combining this reconstruction with automatic differentiation and transposition yields the complexity equivalence. We further transfer Strassen's lower bound for the elementary symmetric functions to the Bezout problem, obtaining an Ω(mlog⁡m)Ω(m\log m) nonscalar lower bound. Hence an O(mlog⁡m)O(m\log m) algorithm, if it exists in this model, would be asymptotically optimal. The reconstruction is genuinely characteristic-dependent: in characteristic pp, all squarefree polynomials Xp+cX+dX^p+cX+d, with fixed c≠0c\ne 0, have the same normalized Bezout pair (0,c−1)(0,c^{-1}). Therefore the all-field, all-input O(mlog⁡m)O(m\log m) problem remains open; generically in characteristic zero, however, it is reduced to the corresponding arbitrary-node multipoint evaluation and interpolation problem.

Figures & tables

Explore similar work

CardsList
  1. Representation Redundancy and Structural Complexity in Finite-Field Inversion

    Sep 4, 2026Zheng Zhang, Na ZhangBoolean Function Learning

  2. Algebraic Expressivity Certificates for Shallow Polynomial Neural Networks

    Sep 11, 2026Sepehr Akbari, Shahrzad JamshidiNeural Network ExpressivityShallow Neural Networks

  3. From Symmetry to Invariance: Learning Galois Equivalent Representations in Finite Fields

    Aug 23, 2026Zheng Zhang, Na ZhangInvariant Representation LearningEquivariant Representation Learning