Presentations of pseudodistributive laws

From MaRDI portal





Distributive laws of monads were introduced by \textit{J. Beck} [Lect. Notes Math. 80, 119--140 (1969; Zbl 0186.02902)], who gave a concise description of the data and coherence conditions needed to compose two monads \(\left( T,u,m\right) \) and \(\left( P,\eta,\mu\right) \), defining a distributive law of monads as a natural transformation \(\lambda:TP\rightarrow PT\) such that the following two triangles and two pentagons commute\N\[\N\begin{array} [c]{ccccc} \NTP & & \xrightarrow{\quad\lambda\quad} & & PT\\\N& _{uP}\nwarrow & & \nearrow_{Pu} & \\\N& & P & & \N\end{array}\N\qquad\qquad\N\begin{array} [c]{ccccc} \NTP & & \xrightarrow{\quad\lambda\quad} & & PT\\\N& _{T\eta}\nwarrow & & \nearrow_{\eta T} & \\\N& & T & & \N\end{array}\N\]\N\[\N\begin{array} [c]{ccccc} \NT^{2}P & \xrightarrow{\ T\lambda\ } &TPT &\xrightarrow{\ \lambda T\ } & PT^{2}\\\N^{mP}\downarrow &&&& \downarrow^{Pm}\\\NTP && \xrightarrow[\quad\lambda\quad]{} && PT \N\end{array}\N\qquad\qquad\N\begin{array} [c]{ccccc} \NTP^{2} & \xrightarrow{\ \lambda P\ } &PTP &\xrightarrow{\ P\lambda\ } & P^{2}T\\\N^{T\mu}\downarrow &&&& \downarrow^{\mu T}\\\NTP && \xrightarrow[\quad\lambda\quad]{} && PT \N\end{array}\N\]\NThis paper considers general 2-dimensional pseudo versions of the pseudodistributive laws. The synopsis of the paper goes as follows.\N\N\begin{itemize}\N\item[\S 2] recalls the monoidal and extensive (no-iteration) definitions of pseudomonads.\N\N\item[\S 3] gives five new presentations of pseudodistributive laws \(\lambda:TP\rightarrow PT\) of pseudomonads:\N\N\begin{itemize}\N\item[(1)] the \textit{pseudomonoidal} definition, which is the pseudo version of Beck's original definition, involving a pseudonatural transformation \(\lambda:TP\rightarrow PT\) and four invertible modifications comprising two triangles and two pentagons abiding by five coherence axioms.\N\N\item[(2)] the \textit{Kleisli-decagon} definition, which involves the decagonal conditions one finds for distributive laws when the involved monads are prsented in extensive form and the usual \(\lambda:TP\rightarrow PT\) is taken as data, comprising three modifications abiding by a version of the triangle and pentagon equations as well as a third comparibility axiom.\N\N\item[(3)] the \textit{pseudoalgebra} definition in terms of maps \(\alpha :TPT\rightarrow PT\), which is a reduced version of the above where a change of variables leads to a simplification in the axioms.\N\N\item[(4)] the \textit{no-iteration} definition in terms of pseudo-pasting operators\N\[\N\left( -\right) ^{\lambda}:\mathcal{C}\left( -,PT-\right) \rightarrow \mathcal{C}\left( T-,PT-\right)\N\]\Nwhich is intended to avoid any iteration of the involved pseudomonads \(T\) and \(P\), which is significant in the relative case [\textit{M. Fiore} et al., Sel. Math., New Ser. 24, No. 3, 2791--2830 (2018; Zbl 1427.18012)].\N\N\item[(5)] the \textit{warping} definition in terms of the data of the Kleisli bicategory of the extended pseudomonad, whose formulation allows for applications of \textit{S. MacLane} and \textit{R. Paré}'s coherence theorem [J. Pure Appl. Algebra 37, 59--80 (1985; Zbl 0567.18003)] and therefore applications to the corresponding pseudodistributive law data in the earlier formulations, generalizing Street and Lack's equivalence of monads and warpings [\textit{S. Lack} and \textit{R. Street}, Cah. Topol. Géom. Différ. Catég. 55, No. 4, 244--266 (2014; Zbl 1349.18012)] to one of distributive laws and distributive warpings.\N\end{itemize}\N\N\item[\S 4] justifies the above five definitions of pseudodistributive law by establishing that they are in equivalence with compatible extensions of \(T\) to the Kleiski bicategory of the pseudomonad \(P\).\N\end{itemize}











This page was built for publication: Presentations of pseudodistributive laws

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7026927)