1
$\begingroup$

Let $X$ be a set equipped with some structure (e.g. topological space, measurable space, probability space, etc.). We say that two endomorphisms $f,g \colon X \to X$ are conjugate to each other if there is an automorphism $h \colon X \to X$ such that $f = h^{-1} \circ g \circ h$; in this case, we refer to $h$ as a conjugacy from $f$ to $g$.

Now suppose the structure on $X$ naturally generates a structure of the same kind on the Cartesian product $X \times X$, with the property that for any two endomorphisms $f$ and $g$ of $X$, the direct product map $f \times g \colon (x_1,x_2) \mapsto (f(x_1),g(x_2))$ is an endomorphism of $X \times X$. We will refer to $f \times f$ as the two-point motion of $f$.

Given an endomorphism $F$ of $X \times X$, a perhaps natural question to ask is whether there exists an endomorphism $f$ of $X$ such that $F$ is conjugate to $f \times f$.

Q1. Is there a line of inquiry within dynamical systems theory and/or ergodic theory that addresses this question, or where this question plays an important role?

Now in the case that $F$ is conjugate to a two-point motion, conjugacies from $F$ to a two-point motion may be very complicated, and not very tractable to work with explicitly. In the general study of conjugacy, one is often not concerned about working explicitly with any conjugacy, but is just concerned with whether a conjugacy exists. However, one property of two-point motions that we might wish to have "explicitly" is the fact that the two components are decoupled from each other.

So we will say that an endomorphism $F$ of $X \times X$ is "nicely conjugate to a two-point motion" if there exist endomorphisms $f,g$ of $X$ that are conjugate to each other and a "tractable" (i.e. not too difficult to work with explicitly) conjugacy from $F$ to $f \times g$.

Q2. Is there a line of inquiry within dynamical systems theory and/or ergodic theory, where endomorphisms of $X \times X$ being "nicely conjugate to a two-point motion" plays an important role?

$\endgroup$
4
  • 1
    $\begingroup$ 'Tractable', 'not too difficult', and 'explicitly' do not seem like defined terms, so it's hard to see how you could prove anything about them (except possibly positive results, where "everyone can tell" that a certain conjugacy is 'tractable'). $\endgroup$
    – LSpice
    Commented Oct 24, 2019 at 16:06
  • 1
    $\begingroup$ For Q2, indeed I'm not thinking in terms of proving results about the general property of being "nicely conjugate to a two-point motion"; I'm not even necessarily thinking in terms of "everyone can tell" results. It seems to me that much of dynamical systems theory concerns not only obtaining final results, but also developing techniques for proving results and/or constructing counterexamples within particular classes of problems. It is from this point of view that I am picturing Q2, with "tractable" referring to the usefulness for whatever broader question is being addressed. $\endgroup$ Commented Oct 24, 2019 at 17:27
  • $\begingroup$ I'm also viewing Q1 from a similar point of view, even though (given the structure of $X$) this can be formulated without the use of soft terms. $\endgroup$ Commented Oct 24, 2019 at 17:34
  • 1
    $\begingroup$ I've now hopefully considerably improved the wording of my questions (especially Q2) to reflect this. Thank you very much, LSpice, for the useful comment. $\endgroup$ Commented Oct 25, 2019 at 16:29

0

You must log in to answer this question.

Browse other questions tagged .