Necklaces, permutations, and periodic critical orbits for quadratic polynomials

Matthew Baker, Andrea Chen, Sophie Li, Matthew Qian · 2025

Read Necklaces, permutations, and periodic critical orbits for quadratic polynomials by Matthew Baker, Andrea Chen, Sophie Li, Matthew Qian (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-12924.txt.

Introduction

Matthew Baker, Andrea Chen, Sophie Li, and Matthew Qian wrote a math paper in 2025 on repeating patterns in a classic map. The map squares a number and adds a constant. For each period length, a polynomial picks out constants where the center point cycles with that exact period.

A past finding linked two counts that seem worlds apart: real roots of that polynomial and factor counts of its even-coefficient twin. The team proves the link by matching sets one to one. Bead necklaces of zeros and ones serve as the bridge between the sets. Each side maps onto the same necklace classes by clear rules. A third match ties the necklaces to cyclic order patterns of a sliding tent map. The proof style is concrete and hands-on.

Catalog record

CREATORMatthew Baker, Andrea Chen, Sophie Li, Matthew Qian
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE85 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-12924

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Necklaces, permutations, and periodic critical orbits for quadratic polynomials Thanks: M.B. thanks Omid Amini, Jason Fulman, Sarah Koch, and Curt McMullen for helpful discussions. A.C, S.L., and M.Q. thank Alon Danai, David Fried, Joseph Kalarickal, Hahn Lheem, John Sim, Leonna Wang, and the PROMYS program for providing them with guidance throughout this project. M.B. was partially supported by NSF grant DMS-2154224 and a Simons Fellowship in Mathematics. Matthew Baker, Andrea Chen, Sophie Li, Matthew Qian Abstract . Let G n G_{n} denote the n th n^{\rm th} Gleason polynomial , whose roots correspond to parameters c c such that the critical point 0 0 is periodic of exact period n n under iteration of z 2 + c z^{2}+c , and let G ¯ n \bar{G}_{n} denote the reduction of G n G_{n} modulo 2 2 . Buff, Floyd, Koch, and Parry made the surprising observation that the number of real roots of G n G_{n} is equal to the number of irreducible factors of G ¯ n \bar{G}_{n} for all n n . We provide a bijective proof for this result by first providing explicit bijections between (a) the set of real roots of G n G_{n} and the set N ¯ ​ ( n ) \bar{N}(n) of equivalence classes of primitive binary necklaces of length n n under the inversion map swapping 0 0 and 1 1 ; and (b) the set of irreducible factors of G n G_{n} modulo 2 and the set N ~ + ​ ( n ) \tilde{N}^{+}(n) of binary necklaces which are either primitive of length n n with an even number of 1 1 ’s or primitive of length n / 2 n/2 with an odd number of 1 1 ’s. We then provide an explicit bijection, closely related to Milnor and Thurston’s kneading theory, between N ¯ ​ ( n ) \bar{N}(n) and N ~ + ​ ( n ) \tilde{N}^{+}(n) . In addition, we provide explicit bijections between N ¯ ​ ( n ) \bar{N}(n) , the set CUP ⁡ ( n ) {\rm CUP}(n) of cyclic unimodal permutations of { 1 , … , n } \{1,\ldots,n\} , and the set N − ​ ( n ) N^{-}(n) of primitive binary necklaces of length n n with an odd number of 1 1 ’s. 1. Introduction For n\geq 1 , let G_{n}(c) denote the n^{\rm th} Gleason polynomial 1 1 1 It is conjectured that G_{n}(c) irreducible over \mathbb{Q} for all n\geq 1 , see [ 2 ] and the references therein. , whose roots correspond to parameters c such that the critical point 0 is periodic of exact period n under iteration of z^{2}+c . For example, G_{3}(c)=\frac{(c^{2}+c)^{2}+c}{G_{1}(c)}=c^{3}+2c^{2}+c+1 and G_{4}(c)=\frac{((c^{2}+c)^{2}+c)^{2}+c}{G_{1}(c)G_{2}(c)}=c^{6}+3c^{5}+3c^{4}+3c^{3}+2c^{2}+1. Let \bar{G}_{n}(c) denote the reduction of G_{n} modulo 2 , considered as an element of {\mathbb{F}}_{2}[c] . Buff et. al. made the surprising observation (cf. [ 2 , Theorem 1.5] ) that the number of real roots of G_{n} is equal to the number of irreducible factors of \bar{G}_{n} for all n\geq 1 . For example, G_{3}(c)=c^{3}+2c^{2}+c+1 has one real root and \bar{G}_{3}(c)=c^{3}+c+1 is irreducible over {\mathbb{F}}_{2} . Similarly, G_{4}(c) has two real roots and \bar{G}_{4}(c)=(c^{4}+c+1)(c^{2}+c+1) . It seems that this remarkable discovery has not yet been adequately “explained”, in the sense that the theorem is proved in [ 2 ] by counting both the number of real roots of G_{n} and the number of irreducible factors modulo 2, and showing that the answer is the same in both cases. The common answer turns out to be \gamma_{n}=\frac{1}{2n}\sum_{m|n,m{\rm\;odd}}\mu(m)2^{\frac{n}{m}} , where \mu is the Möbius function from elementary number theory, characterized by the recursion \sum_{d\mid n}\mu(d)=\left\{\begin{array}[]{ll}1&{\rm if\;}n=1\\ 0&{\rm if\;}n>1.\end{array}\right. The sequence \gamma_{n} is labeled A000048 in the Online Encyclopedia of Integer Sequences. Table 1. The sequence \gamma_{n} n 1 2 3 4 5 6 7 8 9 10 \gamma_{n} 1 1 1 2 3 5 9 16 28 51 It is natural to wonder if there is an explicit bijection between the real roots of G_{n} and the irreducible factors of G_{n} modulo 2. The main result of this paper provides such an explicit bijection. In fact, we do more: we establish explicit bijections between numerous different sets which are all known to have cardinality \gamma_{n} , thus providing a better “reason” why they should have the same cardinality. A foothold into the problem is provided by a famous paper of Milnor and Thurston [ 10 ] , which provides a combinatorial classification of the real roots of G_{n} via kneading theory . In particular, the work of Milnor and Thurston can be used to show that the real roots of G_{n} are canonically in bijection with (a) cyclic unimodal permutations of \{1,\ldots,n\} (cf. [ 5 , Chapter 5] ), (b) I -orbits of periodic cycles of length n for the mod 1 doubling map D:\mathbb{R}/\mathbb{Z}\to\mathbb{R}/\mathbb{Z} sending x to 2x\pmod{1} , where I:x\mapsto-x , and (c) certain kinds of periodic cycles for the tent map T:[0,1/2]\to[0,1/2] , defined as T(x)=2x if 0\leq x\leq 1/4 and T(x)=1-2x if 1/4\leq x\leq 1/2 . Bijection (a) is obtained by recording the way in which iteration of z^{2}+c permutes the elements of the critical orbit, and bijection (b) can be obtained through the spider algorithm [ 8 ] (which is closely related to both the work of Milnor–Thurston [ 10 ] and the Douady–Hubbard theory of external rays [ 6 ] ). Bijection (c) is equivalent to (b) through elementary arguments. Moreover, if we define a necklace to be an equivalence class of binary strings under cyclic rotation, and a necklace inversion class to be an equivalence class of binary necklaces of length n under the involution \iota which swaps 0 and 1 , it is easy to see that inversion classes of periodic cycles of length n for the doubling map D are in bijective correspondence with primitive necklace inversion classes of length n (see Corollary 2.12 ). On the other hand, using an observation of Reutenauer [ 11 ] , it is not difficult (see Section 2.2 for details) to give a bijection between the irreducible factors of \bar{G}_{n} and certain kinds of primitive necklaces of length n . So our task is reduced to the purely combinatorial problem finding explicit bijections between such necklaces and primitive necklace inversion classes. This turns out to be closely connected to Milnor and Thurston’s kneading theory, which we reformulate in a framework that appears to be new. 2 2 2 See [ 3 , Chapter II] and [ 4 , Section 1.18] for related results and additional background on kneading theory. It is interesting to note that although the problem of finding explicit bijections between certain types of necklaces and necklace inversion classes is a purely combinatorial one, making no reference to dynamical systems, our proofs make crucial use of ideas from dynamics. 1.1. Statement of the main result In order to state our main theorem, we need a few definitions. A permutation \sigma of \{1,\ldots,n\} is called cyclic if it consists of a single n -cycle, and unimodal 3 3 3 Some authors require unimodal permutations to increase and then decrease, but the two concepts are equivalent. We note that \sigma is unimodal

Read the complete text