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 501–600 of 2037. Source text is displayed literally.

501\begin{example}[Drift]502For $A=\begin{psmallmatrix}2&1\\1&1\end{psmallmatrix}$ and $x$ the second pure strategy,503$\Q_x(y)=1-(1+q^2)=-q^2<0$ when $y$ plays the first strategy with probability $q>0$. The equilibrium is not504merely not superior; it is inferior to every nearby state. \chk{S4}505\end{example}506 507\section{Properties}508 509The source collects five facts \src{Thms.~1--5}. Each is a statement about $\Q$.510 511\begin{theorem}\label{thm:essprops}512\Scl[Bishop and Cannings 1976; Haigh 1975]513\begin{enumerate}[label=(\alph*)]514\item A game need not have an ESS: Rock--Paper--Scissors has none. \chk{S2}515\item A weakly dominated strategy is not an ESS. \chk{S3}516\item If $x$ is an ESS and $y\ne x$ is a Nash equilibrium, then $\supp y\not\subseteq\supp x$.517\item A game with $n$ strategies has at most $\binom{n}{\lfloor n/2\rfloor}$ ESS.518\item An ESS in the interior of the simplex is the only ESS, and it is globally superior.519\item A quasi-strict equilibrium $x$ is an ESS if and only if $z^{\T}Az<0$ for all $z\ne0$ with520$\supp z\subseteq\supp x$ and $\sum_s z_s=0$.521\end{enumerate}522\end{theorem}523\begin{proof}524(a) By \Cref{cor:uniform} the uniform state $u$ is an equilibrium, and it is the only one. For a pure strategy525$e$, $\pay{u}{e}=0=\pay{e}{e}$ and $\pay{e}{u}=0=\pay{u}{u}$, so both clauses of (i) fail.526(b) If $\mu$ weakly dominates $x$ then $\pay{\mu}{x}\ge\pay{x}{x}$ and $\pay{\mu}{\mu}\ge\pay{x}{\mu}$, so527$\Q_x(\mu)\le0$ with $\mu$ an alternative best reply.528(c) If $\supp y\subseteq\supp x$ then $y$ is a best reply to $x$, since every pure strategy in the support of529$x$ is. The second clause of (i) gives $\pay{x}{y}>\pay{y}{y}$, contradicting that $y$ is an equilibrium.530(d) By (c) the supports of distinct ESS form an antichain of subsets of $S$; apply Sperner's theorem.531(e) If $x$ is interior, every $y$ is a best reply to $x$, so $\Q_x(y)>0$ for all $y\ne x$; then no other state532is an equilibrium by (c).533(f) For $y$ near $x$ write $y-x=z+w$ with $z$ supported in $\supp x$; quasi-strictness makes the first-order534term in $w$ strictly positive, and for $w=0$, $\Q_x(y)=-z^{\T}Az$ because $z^{\T}Ax=0$.535\end{proof}536 537Criterion (f) is the one used by the companion script to decide evolutionary stability exactly.538 539\section{Sets, and why they are forced}\label{sec:essets}540 541A population may be stable as a whole while drifting inside a set of equivalent states. The source surveys542Thomas's evolutionarily stable sets and three variants due to Balkenborg and Schlag \src{\S2.3}. We use the543following form.544 545\begin{definition}\label{def:esset}546\Scl[Thomas 1985; \cite{Wei95}] A non-empty closed set $L\subseteq\Dist(S)$ is an \term{evolutionarily stable547set} if every $x\in L$ has a neighbourhood $U$ with $\Q_x(y)\ge0$ for all $y\in U$, and $\Q_x(y)>0$ for548$y\in U\setminus L$.549\end{definition}550 551\noindent A one-point set $\{x\}$ is an evolutionarily stable set exactly when $x$ is an ESS, and every point552of an evolutionarily stable set is a Nash equilibrium.553 554\begin{theorem}[Stability is a property of sets]\label{thm:esset}555\Sob Let $f\colon(S,A)\to(T,B)$ be a payoff-preserving surjection and let $L\subseteq\Dist(T)$ be an evolutionarily556stable set. Then $f_*^{-1}(L)$ is an evolutionarily stable set of $(S,A)$. If $f$ is not injective and $x$ is557an ESS of $(T,B)$ whose support meets the set of duplicated strategies, then no point of $f_*^{-1}(x)$ is an558ESS.559\end{theorem}560\begin{proof}561By \Cref{prop:push}, $\Q_\sigma(\mu)=\Q_{f_*\sigma}(f_*\mu)$ for all $\sigma,\mu\in\Dist(S)$. Let562$\sigma\in f_*^{-1}(L)$ and let $U$ be the neighbourhood of $f_*\sigma$ from \Cref{def:esset}. Then563$f_*^{-1}(U)$ is a neighbourhood of $\sigma$ on which $\Q_\sigma\ge0$, with strict inequality off564$f_*^{-1}(L)$; and $f_*^{-1}(L)$ is closed and non-empty. For the second statement, the fibre over $x$ contains565a segment through each of its points, on which $\Q_\sigma$ vanishes identically.566\end{proof}567 568\begin{example}569Duplicate Dove in the Hawk--Dove game. The fibre over the ESS is the segment570$L=\{(p,s,1-p-s)\}$, and $\Q_\sigma(\mu)=\tfrac C2(p-\mu_{\mathrm H})^2$ for $\sigma\in L$: zero exactly on $L$,571positive elsewhere. \chk{S6}572\end{example}573 574\Cref{thm:esset} explains a development that the source records historically. Duplicating a strategy changes575nothing about the biology or the sociology of a model; it is a change of description. A solution concept576that is destroyed by it cannot be the right concept, and the least repair is to pass from points to sets.577The three variants of Balkenborg and Schlag differ in the order of quantifiers over the neighbourhood (one578threshold for each invader, for each incumbent, or for all) and so in how they behave when $L$ is not compact579in the relevant topology; for the closed sets of \Cref{def:esset} we do not need the distinction.580 581\section{Relative entropy}582 583The invasion form has a second life. For $x,y\in\Dist(S)$ with $\supp x\subseteq\supp y$ let584\[ \KL{x}{y}=\sum_{s}x_s\log\frac{x_s}{y_s} \]585be the relative entropy. It is the quantity singled out in \cite{ESCHC}, following Baez and Fritz, as the586unique functorial measure of information lost by a stochastic map. In \Cref{ch:replicator} we shall see that587along the replicator dynamics588\[ \frac{d}{dt}\KL{x}{y(t)}=-\Q_x(y(t)), \]589so that local superiority of $x$ says precisely that relative entropy to $x$ decreases strictly along nearby590trajectories. Evolutionary stability is the statement that a population near $x$ \emph{gains information591about} $x$.592 593\section*{Notes}594 595The definition of an ESS is from \cite{MSP73}; the equivalences of \Cref{thm:essequiv} are in \cite{HSS79} and596\cite[Ch.~2]{Wei95}. Parts (c)--(e) of \Cref{thm:essprops} are due to Bishop and Cannings \cite{BiCa76}, part (f)597to Haigh \cite{Hai75}; the sharp maximal number of ESS for given $n$ is a harder problem than the Sperner598bound suggests. Evolutionarily stable sets were introduced by Thomas \cite{Tho85} and refined in599\cite{BaSc01}. The observation that they are the closure of the point concept under duplication of600strategies is, as far as we know, not stated in this form in the literature, though it is implicit in the