ProbXiv
Explorersign in
Problem archiveProblem record

Statement

Let n≥2n \ge 2 and let ff be a smooth dynamical system on Rn\mathbb{R}^n (a diffeomorphism, or the time-one map of a smooth flow) possessing a compact attractor AA that carries an ergodic natural measure μ\mu, i.e. a probability measure supported on AA such that for Lebesgue-almost every initial point xx in the basin of attraction of AA the orbit time averages converge to the μ\mu-averages,

lim⁡N→∞1N∑k=0N−1φ(fk(x))  =  ∫φ dμfor every continuous φ.\lim_{N \to \infty} \frac{1}{N} \sum_{k=0}^{N-1} \varphi\big(f^{k}(x)\big) \;=\; \int \varphi \, d\mu \qquad \text{for every continuous } \varphi .

Let λ1≥λ2≥⋯≥λn\lambda_{1} \ge \lambda_{2} \ge \dots \ge \lambda_{n} be the Lyapunov exponents of ff with respect to μ\mu, listed with multiplicity, i.e. the exponential rates of growth

λi  =  lim⁡N→∞1Nlog⁡σi ⁣(DfN(x)),\lambda_{i} \;=\; \lim_{N \to \infty} \frac{1}{N} \log \sigma_{i}\!\left( Df^{N}(x) \right),

where σ1≥⋯≥σn\sigma_{1} \ge \dots \ge \sigma_{n} denote singular values and the limit exists for μ\mu-almost every xx.

Assume λ1≥0\lambda_{1} \ge 0 and ∑i=1nλi<0\sum_{i=1}^{n} \lambda_{i} < 0, and let j∈{1,…,n−1}j \in \{1, \dots, n-1\} be the largest index with

∑i=1jλi≥0,∑i=1j+1λi<0,\sum_{i=1}^{j} \lambda_{i} \ge 0, \qquad \sum_{i=1}^{j+1} \lambda_{i} < 0 ,

so that λj+1<0\lambda_{j+1} < 0. Define the Kaplan–Yorke (Lyapunov) dimension

DKY  =  j+∑i=1jλi∣λj+1∣.D_{KY} \;=\; j + \frac{\sum_{i=1}^{j} \lambda_{i}}{\left| \lambda_{j+1} \right|} .

Define the information dimension of μ\mu as follows: for ε>0\varepsilon > 0 let {Ci(ε)}i\{ C_{i}(\varepsilon) \}_{i} be the cubes of side ε\varepsilon of a grid partition of Rn\mathbb{R}^n, put μi(ε)=μ(Ci(ε))\mu_{i}(\varepsilon) = \mu\big( C_{i}(\varepsilon) \big) and

H(ε)  =  −∑i : μi(ε)>0μi(ε)log⁡μi(ε),DI  =  lim⁡ε→0+H(ε)log⁡(1/ε).H(\varepsilon) \;=\; - \sum_{i \,:\, \mu_{i}(\varepsilon) > 0} \mu_{i}(\varepsilon) \log \mu_{i}(\varepsilon), \qquad D_{I} \;=\; \lim_{\varepsilon \to 0^{+}} \frac{H(\varepsilon)}{\log (1/\varepsilon)} .

Then for a typical such system the limit DID_{I} exists and

DI  =  DKY  =  j+∑i=1jλi∣λj+1∣;D_{I} \;=\; D_{KY} \;=\; j + \frac{\sum_{i=1}^{j} \lambda_{i}}{\left| \lambda_{j+1} \right|} ;

that is, the dimension of the attractor AA equals DKYD_{KY}.

Record

Added
Links

Comments

Nothing has been published against this problem yet, and nothing has been said about it.

Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.

Sign in

Solve with an agent

Open the statement in a chat, with the problem and the ground rules already written into the prompt.

This opens a third-party site. Nothing is posted back to ProbXiv and nothing you write there is recorded here — what a model gives you is an attempt, which a person still has to check.