OctoInformBeta

OpenAI preprint, September 2026 · §1. Introduction

Theorem 1.1: Self-similar probability measures with exact overlaps and repeated generators

Dimension from the entropy of affine compositions.

Hover over a blue or green term in the text, or a node in the graph, to find it in the other. Select it to read its declaration both in Lean and in words.

The Setting

A finite family of contracting similarities on the line generates words whose affine maps may coincide. The relevant random walk records the complete affine map reached by a word, including its signed slope and its translation.

A weighted affine system (φi,pi)i∈I(\varphi_i,p_i)_{i\in I} on the line is a family of similarities φi\varphi_i with strictly positive probabilities pip_i, indexed by a finite nonempty set II and satisfying (1), where ri,ti∈ℝr_i,t_i\in\mathbb R: φi(x)=rix+ti,0<|ri|<1,pi>0,∑i∈Ipi=1.\varphi_i(x)=r_i x+t_i,\qquad 0<|r_i|<1,\qquad p_i>0,\qquad \sum_{i\in I}p_i=1.(1) A self-similar probability measure μ\mu for a weighted affine system (φi,pi)i∈I(\varphi_i,p_i)_{i\in I} on the line is a probability measure on ℝ\mathbb R satisfying the invariance equation (2), in which (φi)*μ(\varphi_i)_*\mu denotes the pushforward of μ\mu by φi\varphi_i: μ=∑i∈Ipi(φi)*μ.\mu=\sum_{i\in I}p_i(\varphi_i)_*\mu.(2)

The Affine Walk

The complete affine map gwg_w of a finite word w=(i1,…,in)∈Inw=(i_1,\ldots,i_n)\in I^n is the pair (rw,tw)(r_w,t_w) determined by the composition in (3); the word probability pwp_w is the product given in the same display, and ⌀\varnothing denotes the empty word: φw=φi1∘⋯∘φin,φw(x)=rwx+tw,g⌀=(1,0),pw=∏j=1npij.\varphi_w=\varphi_{i_1}\circ\cdots\circ\varphi_{i_n},\qquad \varphi_w(x)=r_wx+t_w,\qquad g_{\varnothing}=(1,0),\qquad p_w=\prod_{j=1}^{n}p_{i_j}.(3) The attainable maps at depth nn form the finite set AnA_n in (4); the random walk mass qn(g)q_n(g) is the probability that a word of length nn has complete affine map gg: An={gw:w∈In},qn(g)=∑w∈Ingw=gpw.A_n=\{g_w:w\in I^n\},\qquad q_n(g)=\sum_{\substack{w\in I^n\\g_w=g}}p_w.(4)

The map-walk entropy HnH_n is the Shannon entropy of qnq_n, and its entropy rate hh is the infimum of the entropies per letter over positive word lengths, as in (5). The logarithms are to base two, with 0log⁡20=00\log_2 0=0. Hn=−∑g∈Anqn(g)log⁡2qn(g),h=infn≥1Hnn.H_n=-\sum_{g\in A_n}q_n(g)\log_2 q_n(g),\qquad h=\inf_{n\geq 1}\frac{H_n}{n}.(5) The positive Lyapunov exponent χ\chi of a weighted affine system (φi,pi)i∈I(\varphi_i,p_i)_{i\in I} on the line is its mean logarithmic contraction per letter, defined by (6) with logarithms to base two: χ=−∑i∈Ipilog⁡2|ri|.\chi=-\sum_{i\in I}p_i\log_2|r_i|.(6)

The lower Hausdorff dimension dim⁡Hμ\dim_H\mu of a measure μ\mu on ℝ\mathbb R is the extended nonnegative real number given by (7), where dim⁡HE\dim_H E denotes the Hausdorff dimension of a set EE: dim⁡Hμ=inf⁡{dim⁡HE:E⊆ℝ is Borel and μ(E)>0}.\dim_H\mu= \inf\{\dim_H E:E\subseteq\mathbb R\text{ is Borel and }\mu(E)>0\}.(7)✦

The Claim

Theorem 1. Let (φi,pi)i∈I(\varphi_i,p_i)_{i\in I} be a weighted affine system satisfying (1), and let μ\mu be a self-similar probability measure satisfying (2). For the entropy rate hh and Lyapunov exponent χ\chi defined in (5) and (6), its lower Hausdorff dimension is given by (8): dim⁡Hμ=min⁡{1,hχ}.\dim_H\mu=\min\left\{1,\frac{h}{\chi}\right\}.(8)✦

What Is Checked

The claim is stated over every finite nonempty weighted contracting affine system on the line and every probability measure satisfying its self-similarity equation.