Independent scholarship · Open accessThe monthly digest

Indexed text attachment

evolutionary-game-theory-in-higher-categories.tex

Source for Evolutionary Game Theory in Higher Categories

View paper · Download original file

Lines 377–476 of 2037. Source text is displayed literally.

377 378A column shift changes $\pay{x}{y}$ by $c^{\T}y$, which does not depend on $x$; it changes no comparison379between strategies. Every game is a column shift of a unique game with zero diagonal, and the source's380normalization to positive payoffs \src{\S4.1} is a column shift by a constant vector.381 382\section{Equilibria}383 384\begin{definition}385A state $x\in\Dist(S)$ is a \term{(symmetric) Nash equilibrium} of $(S,A)$ if $\pay{x}{x}\ge\pay{y}{x}$ for all386$y\in\Dist(S)$; equivalently, if $(Ax)_s\le x^{\T}Ax$ for all $s$, with equality for $s\in\supp x$. It is387\term{strict} if $\pay{x}{x}>\pay{y}{x}$ for all $y\ne x$, and \term{quasi-strict} if $(Ax)_s<x^{\T}Ax$ for all388$s\notin\supp x$.389\end{definition}390 391\begin{theorem}\label{thm:nash}392\Scl[Nash 1951] Every symmetric game has a symmetric Nash equilibrium. If a finite group $G$ acts on $(S,A)$ by393symmetries, the set of Nash equilibria is $G$-invariant and contains a $G$-fixed point.394\end{theorem}395 396\begin{corollary}\label{cor:uniform}397If a group of symmetries acts transitively on $S$, the uniform state is a Nash equilibrium. In particular the398uniform state is an equilibrium of every game of \Cref{ex:rps}.399\end{corollary}400 401How equilibria behave under the maps of \Cref{def:map} is recorded once and used repeatedly.402 403\begin{proposition}\label{prop:nashmaps}404\Sob Let $f\colon(S,A)\to(T,B)$ be payoff-preserving.405\begin{enumerate}[label=(\alph*)]406\item If $f$ is surjective, then $x$ is a Nash equilibrium of $(S,A)$ if and only if $f_*x$ is a Nash407equilibrium of $(T,B)$.408\item If $f$ is injective, then $f_*x$ Nash in $(T,B)$ implies $x$ Nash in $(S,A)$; the converse holds if and409only if no strategy outside the image of $f$ earns more than $\pay{x}{x}$ against $f_*x$.410\end{enumerate}411\end{proposition}412\begin{proof}413By \Cref{prop:push} the pure payoffs against $f_*x$ at strategies in the image of $f$ are the pure payoffs414against $x$, and the mean payoffs agree. The equilibrium condition on $T$ is the equilibrium condition on $S$415together with a condition at strategies outside the image, which is vacuous when $f$ is surjective.416\end{proof}417 418The asymmetry between (a) and (b) recurs at every level of the theory: clone maps are invisible to selection,419inclusions are visible exactly through the strategies they omit.420 421\section*{Notes}422 423The distribution monad and convex spaces as its algebras go back to Stone and to \'Swirszcz; a modern account424is in \cite{Fri09}. \Cref{thm:nash} is from Nash's paper on non-cooperative games \cite{Nas51}. The category425$\Game$ is not standard. Categories of games with strategy-changing morphisms appear in the literature on426potential games and on open games \cite{GHWZ18}; those morphisms are richer and are not needed here. The427classical theory with rational players is the subject of \cite{GIHC}. The invariance of evolutionary dynamics428under column shifts is used systematically in \cite{Zee80,HoSi98}.429 430\section*{Exercises}431 432\begin{exercise}\label{exer:biaffine}433Let $y_\epsilon=(1-\epsilon)x+\epsilon\mu$. Show that434$\pay{x}{y_\epsilon}-\pay{y_\epsilon}{y_\epsilon}=\epsilon\,\bigl(\pay{x}{y_\epsilon}-\pay{\mu}{y_\epsilon}\bigr)$.435\end{exercise}436 437\begin{exercise}\label{exer:aut}438Determine the group of symmetries of Rock--Paper--Scissors and of the five-strategy game of \Cref{ex:rps}.439\end{exercise}440 441\begin{exercise}\label{exer:hd}442For the Hawk--Dove game with $A=\begin{psmallmatrix}(V-C)/2&V\\0&V/2\end{psmallmatrix}$, find the zero-diagonal443normal form under column shifts and read off the equilibria for $V<C$ and for $V>C$.444\end{exercise}445 446\begin{exercise}\label{exer:clone}447Let $(S,A)$ be the Hawk--Dove game with the strategy Dove duplicated. Describe all its Nash equilibria for448$V<C$.449\end{exercise}450%% ---- end ch/01-games.tex ----451%% ---- begin ch/02-stability.tex ----452\chapter{Evolutionary Stability}\label{ch:stability}453 454\section{The invasion form}455 456A Nash equilibrium asks that no deviation does better against the incumbent. In a population that is too457weak: a deviation that does equally well against the incumbent is not selected against, and may spread if it458does better against itself. The source's first example \src{\S2.1} is the game459$\begin{psmallmatrix}2&1\\1&1\end{psmallmatrix}$, in which the second strategy is an equilibrium that the first460invades by drift. All of the static theory can be phrased with one function.461 462\begin{definition}\label{def:Q}463The \term{invasion form} of a state $x$ is the function on $\Dist(S)$464\[ \Q_x(y)=\pay{x}{y}-\pay{y}{y}, \]465the advantage of $x$ over $y$ in a population in state $y$.466\end{definition}467 468\begin{definition}\label{def:ess}469\Scl[Maynard Smith and Price 1973; \src{Defs.~1--4}] A state $x$ is470\begin{enumerate}[label=(\roman*)]471\item an \term{evolutionarily stable strategy} (ESS) if for every $\mu\ne x$ either $\pay{x}{x}>\pay{\mu}{x}$, or472$\pay{x}{x}=\pay{\mu}{x}$ and $\pay{x}{\mu}>\pay{\mu}{\mu}$;473\item \term{locally superior} if $\Q_x(y)>0$ for all $y\ne x$ in a neighbourhood of $x$;474\item protected by a \term{uniform invasion barrier} if there is $\bar\epsilon>0$ such that475$\pay{x}{y_\epsilon}>\pay{\mu}{y_\epsilon}$ for all $\mu\ne x$ and $0<\epsilon<\bar\epsilon$, where476$y_\epsilon=(1-\epsilon)x+\epsilon\mu$.