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 761–860 of 2037. Source text is displayed literally.

761\end{proof}762 763\noindent When $s$ is a twin of another strategy $s'$ with every payoff lowered by $\delta$, the proof gives an764identity: $\log(x_s/x_{s'})$ decreases at the constant rate $\delta$. \chk{R9} The hypothesis that the solution be765interior is the source's proviso that all strategies be initially present, and by \Cref{cor:three}(a) it766cannot be dropped.767 768\section{Cycles}769 770\begin{theorem}\label{thm:zerosum}771\Scl[\cite{HoSi98}] Let $A$ be antisymmetric with an interior equilibrium $x^*$. Then $\KL{x^*}{y}$ is constant772along every interior solution of \eqref{eq:rep}. No interior solution converges to $x^*$ or to the boundary.773\end{theorem}774\begin{proof}775$y^{\T}Ay=0$ and $x^{*\T}Ay=-y^{\T}Ax^*=0$, so $\Q_{x^*}\equiv0$; apply \Cref{thm:kl}.776\end{proof}777 778For Rock--Paper--Scissors this gives the family of closed orbits around the centre. \chk{R4}779 780\begin{proposition}[Generalized Rock--Paper--Scissors]\label{prop:grps}781\Scl[\cite{Zee80}] Let a win pay $a>0$ and a loss $-b<0$. The linearization of \eqref{eq:rep} at the uniform782state has eigenvalues783\[ \lambda_\pm=\frac{b-a}{6}\pm i\,\frac{\sqrt3\,(a+b)}{6}. \]784The centre is asymptotically stable and an ESS for $a>b$, a centre for $a=b$, and a repeller for $a<b$, when785interior orbits approach the boundary. \chk{R5, R6}786\end{proposition}787 788\noindent The cases $a>b$ and $a<b$ are the source's ``good'' and ``bad'' versions. By \Cref{cor:three}(c) the789uniform state is a rest point for every $a,b$, being the fixed point of the $\Z/3$ of symmetries. For three790strategies the possible phase portraits were classified by Zeeman and by Bomze \cite{Zee80,Bom83}; they are791assembled from rest points and cycles. With four strategies the state space is three-dimensional and792chaotic attractors occur \cite{ASS80}; the source illustrates one \src{\S3.2.5}. Two statements survive chaos.793 794\begin{theorem}\label{thm:average}795\Scl[\cite{HoSi98}] If a solution of \eqref{eq:rep} stays at a positive distance from the boundary, then every796limit point of its time averages $\frac1T\int_0^Tx(t)\,dt$ is an interior equilibrium. If the interior797equilibrium is unique, the time averages converge to it.798\end{theorem}799 800\noindent The second surviving statement is the index theorem of the next chapter.801 802\section*{Notes}803 804The replicator equation was introduced in \cite{TaJo78} and named in \cite{ScSi83}. The standard reference is805\cite{HoSi98}; \cite{San10} treats it within a general theory of population dynamics. \Cref{thm:gradient} is in806\cite{Sha79,Aki79}. The functorial statement \Cref{thm:natural} is elementary and its three cases are807classical; we have not found it stated as a single naturality property. Relative entropy as a Lyapunov808function, and the reading of the replicator equation as an information-geometric flow, are developed in809\cite{BaPo16,Har09}. The equivalence of the replicator equation on $n$ strategies with a Lotka--Volterra810system in $n-1$ variables is due to Hofbauer \cite{Hof81}.811 812\section*{Exercises}813 814\begin{exercise}\label{exer:two}815Derive \eqref{eq:two} and the escape-time formula for the drift game.816\end{exercise}817 818\begin{exercise}\label{exer:fisher}819Prove the formula for $\frac{d}{dt}x^{\T}Ax$ in \Cref{thm:gradient}, and show that for general $A$ the rate is820$2\operatorname{Cov}_x(Sx,Ax)$, where $S=\tfrac12(A+A^{\T})$.821\end{exercise}822 823\begin{exercise}\label{exer:eig}824Verify \Cref{prop:grps} by computing the trace and determinant of the linearization on the tangent space.825\end{exercise}826 827\begin{exercise}\label{exer:lump}828A population plays Rock--Paper--Scissors with Rock present in two indistinguishable variants, in initial829state $(\tfrac14,\tfrac14,\tfrac14,\tfrac14)$. Describe the motion of the class frequencies.830\end{exercise}831%% ---- end ch/03-replicator.tex ----832%% ---- begin ch/04-index.tex ----833\chapter{Index, Recurrence, and Decomposition}\label{ch:index}834 835Three questions cannot be answered one equilibrium at a time. How many equilibria can there be, and of what836kinds? What replaces an equilibrium when the dynamics cycles? Which part of a game is responsible for the837cycling? Each has an answer in which topology does the work.838 839\section{The index theorem}840 841Write $F_A$ for the replicator field. At a rest point $x$ its derivative preserves the tangent space842$T=\{v:\sum_sv_s=0\}$ of the simplex.843 844\begin{definition}845A game is \term{regular} if every Nash equilibrium is quasi-strict and $D_xF_A|_T$ is invertible at each. The846\term{index} of a Nash equilibrium $x$ of a regular game is847\[ \operatorname{ind}(x)=\operatorname{sign}\det\bigl(-D_xF_A|_T\bigr)\in\{+1,-1\}. \]848\end{definition}849 850\noindent Regular games are open and dense among all games. An equilibrium at which every eigenvalue of851$D_xF_A|_T$ has negative real part has index $+1$.852 853\begin{theorem}[Index theorem]\label{thm:index}854\Scl[Hofbauer and Sigmund; Poincar\'e--Hopf] For a regular game,855\[ \sum_{x\ \mathrm{Nash}}\operatorname{ind}(x)=1 . \]856\end{theorem}857\begin{proof}[Sketch]858For small $\epsilon>0$ the field $F_A(x)+\epsilon(u-x)$, with $u$ the uniform state, points strictly inward859on the boundary of the simplex. Its rest points converge, as $\epsilon\to0$, to the rest points of $F_A$ that860are \emph{saturated}, that is, to the Nash equilibria: a rest point with a strategy outside its support that