Talk:Leavitt path algebra

Table

I think that the first row of the table linking the algebraic properties for LPA's with properties of the graph is incorrect. There are many examples of LPAs (for example the Laurent polynomials arising from a single vertex and loop) which are infinite dimensional (as $K$-vector spaces). Maybe you were looking for finite and acyclic iff von neumann regular (which is similar to "finite" in graph $C^*$-algebras)? — Preceding unsigned comment added by 2603:8081:7F00:CB6E:CD17:6391:6C5F:418A (talk) 02:00, 25 January 2022 (UTC)

Unnreliable?

I suspect the definition of a simple cycle is incorrect. It is completely different from any other definition I've seen and does not correspond to the idea of a simple closed curve. Are there other errors? Am I all wrong? Zaslav (talk) 08:17, 30 August 2023 (UTC)

Hi Zaslav -- I believe the definition is correct. This is the most commonly used definition for a "simple cycle"; i.e., that the directed path returns to its base only at the end (but can repeat intermediate vertices on the way). The terminology and definition come from the subject of graph C*-algebras, and it works particularly well for describing "Condition (K)", an important condition that arises in the study of both graph C*-algebras and Leavitt path algebras (LPAs). In both graph C*-algebras and LPAs it's been acknowledged that the most useful graph concepts in these subjects often don't align with concepts encountered in directed graph theory or its adjacent subjects. Graph C*-algebraists (working primarily in functional analysis) have been happy to make up their own terms and even have terminology conflicting with directed graph theory, and unfortunately this terminology has become standard among graph C*-algebraists. (The term "loop" is a prime example of this.) People working in LPAs have made more effort to align their terminology with graph theorists, but there is still incongruity. As far as I know, the article's definition of "simple cycle" agrees with the existing literature on LPAs and is the most common definition used by those working in LPAs. But if you have recent articles showing a shift in terminology, I'd be happy to discuss the issue further and continue a discussion on if/how the definition in this article should be modified. (Also: Thank you for your other minor edits to improve the article. They are much appreciated.) QuietSisyphus (talk) 21:48, 30 August 2023 (UTC)