The convergence of a sequence of polynomials and the distribution of their zeros

Min-Hee Kim, Young-One Kim, Jungseob LeearXiv 2019 · DOI · arXiv

Abstract: Suppose that $\langle f_n \rangle$ is a sequence of polynomials, $\langle f_n^{(k)}(0)\rangle$ converges for every non-negative integer $k$, and that the limit is not $0$ for some $k$. It is shown that if all the zeros of $f_1, f_2, \dots$ lie in the closed upper half plane $\rm{Im}\ z\geq 0$, or if $f_1, f_2, \dots$ are real polynomials and the numbers of their non-real zeros are uniformly bounded, then the sequence converges uniformly on compact sets in the complex plane. The results imply a theorem of Benz and a conjecture of Pólya.

Introduction

In the study of complex analysis, a fundamental question concerns the relationship between the local behavior of a function at a point and its global behavior across the complex plane. For sequences of polynomials, this often manifests as the distinction between weak convergence (where the coefficients of the Taylor series, or derivatives at the origin, converge) and strong convergence (where the functions converge uniformly on any compact set).

Generally, weak convergence does not imply strong convergence. A sequence of polynomials can have coefficients that stabilize, yet the functions themselves may exhibit divergent or chaotic behavior as one moves away from the origin. However, a long-standing intuition in the field suggests that if the zeros of these polynomials are restricted to specific geometric regions, the system becomes "rigid." This paper provides a definitive answer to this century-old problem, proving that specific constraints on zero distribution force weak convergence to become strong convergence.

The Problem of Geometric Rigidity

The core of this research addresses whether the location of a polynomial's roots can dictate its convergence properties. Mathematicians have long suspected that if zeros are forbidden from certain regions of the complex plane, the function loses the freedom to diverge if its behavior at the origin is fixed.

This paper focuses on two primary geometric constraints:

  1. Zeros located entirely within the closed upper half-plane ($\text{Im } z \geq 0$).
  2. Real polynomials where the number of non-real zeros is uniformly bounded across the sequence.

By establishing convergence under these conditions, the authors resolve a conjecture made by George Pólya in 1915 and complete a theoretical framework initiated by Benz in 1934.

Methodology and Mathematical Strategy

The authors employ several sophisticated strategies to bridge the gap between local and global convergence:

  • Generalizing Classical Theorems: The research builds upon the 1914 Lindwart-Pólya theorem. While the original theorem required the sequence to converge uniformly on a small disk, the authors demonstrate that mere coefficient convergence is sufficient when zeros are appropriately constrained.
  • Reciprocal Power Sums: By analyzing the sums of the reciprocal powers of the zeros, denoted as $s_k(f)$, the authors establish a link between the distribution of zeros and the derivatives of the function at zero. Controlling these sums allows for control over the growth of the function.
  • Jensen and Appell Polynomials: These specific polynomial types serve as a bridge to show that approximating polynomials with zeros in the upper half-plane must converge to a limit function within the Pólya-Obrechkoff ($PO$) class.
  • Topological Rigidity: For real polynomials with a bounded number of non-real zeros, the authors utilize arguments involving compact topological groups to decompose the sequence into finite, strongly convergent subsequences.

Key Results

The paper presents two landmark theorems that define the "weak-to-strong" leap:

Theorem 1.1: Upper Half-Plane Constraints

If a sequence of polynomials with zeros in the closed upper half-plane converges weakly (and the limit is not identically zero), it must converge strongly to an entire function in the Pólya-Obrechkoff class. This result provides the missing link in Benz’s 1934 work, which identified the limit class but could not prove uniform convergence.

Theorem 1.2: Bounded Non-Real Zeros

If the polynomials are real and the number of their non-real zeros is uniformly bounded, weak convergence implies strong convergence to a function in the $LP^*$ class (an extension of the Laguerre-Pólya class).

Resolution of Pólya’s Conjecture

Applying these theorems to Appell and Jensen polynomials, the authors successfully prove Pólya’s 1915 conjecture. They confirm that if the number of non-real zeros of these polynomials remains bounded, the underlying power series represents a specific type of entire function characterized by its zero distribution.

Conclusion and Implications

This work demonstrates that the geometry of zeros acts as a powerful regularizing force in complex analysis. The requirement that at least one coefficient of the limit function be non-zero remains a critical boundary condition; without it, sequences like $f_n(z) = z^n$ provide simple counterexamples where coefficients vanish while the function diverges.

While this research settles the convergence criteria for the upper half-plane and real lines with bounded exceptions, it opens new avenues for exploring whether similar rigidity exists for more complex or exotic regions of the complex plane. Furthermore, the qualitative proof of convergence invites future quantitative studies into the specific rates of convergence relative to zero density.

Summary updated

Abstract

Suppose that $\langle f_n \rangle$ is a sequence of polynomials, $\langle f_n^{(k)}(0)\rangle$ converges for every non-negative integer $k$, and that the limit is not $0$ for some $k$. It is shown that if all the zeros of $f_1, f_2, \dots$ lie in the closed upper half plane $\rm{Im}\ z\geq 0$, or if $f_1, f_2, \dots$ are real polynomials and the numbers of their non-real zeros are uniformly bounded, then the sequence converges uniformly on compact sets in the complex plane. The results imply a theorem of Benz and a conjecture of Pólya.
BibTeX
@misc{kim2019convergence,
  author = {Min-Hee Kim and Young-One Kim and Jungseob Lee},
  title = {{The convergence of a sequence of polynomials and the distribution of their zeros}},
  year = {2019},
  eprint = {1903.01140},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1903.01140},
}
Figure