On the most expected number of components for random links (Q1705558)

From MaRDI portal
scientific article
Language Label Description Also known as
English
On the most expected number of components for random links
scientific article

    Statements

    On the most expected number of components for random links (English)
    0 references
    0 references
    0 references
    16 March 2018
    0 references
    In [\textit{J. Ma}, J. Knot Theory Ramifications 22, No. 8, Article ID 1350043, 11 p. (2013; Zbl 1278.57012)] two models of random links were introduced. The authors of the present paper consider the braid closures of randomly chosen braids obtained via random walks on the braid groups. Suppose that such a random walk on the braid group \({\mathcal B}_n\) of \(n\)-strings induces the uniform distribution on the symmetric group \(\sigma_n\) on \(n\) letters via the natural projection \({\mathcal B}_n\to \sigma_n\), \(n \geq 3\). Ma showed in [loc. cit.] that for the random link coming from a random walk of \(k\)-steps on \({\mathcal B}_n\), the expected value of the number of components converges to \(\sum^n_{k=1}{{1}\over{k}}\) when \(k\) goes to \(\infty\). The main results of the paper are the following theorems. Theorem 1. Consider a random link obtained from a random walk on \({\mathcal B}_n\). Suppose that the random walk on \({\mathcal B}_n\) is defined for the probability distribution on \({\mathcal B}_n\) which induces the uniform distribution on \(\sigma_n\) via the natural projection \({\mathcal B}_n\to \sigma_n\), \(n \geq 3\). Then the most expected number of components is equal to \(K_n = \left[\log(n+1) + \gamma -1 + {{\zeta(2) - \zeta(3)}\over {\log(n+1) + \gamma - 1.5}}+ {{h}\over{(\log(n+1) + \gamma - 1.5)^2}}\right]\), where \([x]\) denotes the integer part of \(x\), \(\zeta\) is the Riemann zeta function, \(\gamma= 0.5772\dots\) is the Euler-Mascheroni constant, and \(h\) with \(-1.1< h< 1.5\) is a function on \(n\). In particular, if \(n>188\), it follows that \(\left[\log n - 1/2\right]<K_n < \left[\log n\right]\). Theorem 2. Consider a random braid obtained from a random walk on \({\mathcal B}_n\). Suppose that the random walk on \({\mathcal B}_n\) is defined for the probability distribution on \({\mathcal B}_n\) which induces the uniform distribution on \(\sigma_n\) via the natural projection \({\mathcal B}_n\to \sigma_n\), \(n \geq 3\). Then the most expected partition of the number of the strings is \(((n-1), 1)\).
    0 references
    random link
    0 references
    random walk
    0 references
    braid
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references