Jump to content

Talk:Evolutionary game theory

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

EGT != Biological game theory

[edit]

I edited the page to remove the text which seemed to be equating evolutionary game theory' to game theory as practised by biologists, and refocussed on the definition of EGT as game theory with dyamical systems. Text removed placed below. I also removed the text about not behaving rationally, since biological game theory models (e.g. Hawk Dove, etc) don't assume rationality, and yet aren't EGT models. Pete.Hurd 04:29, 1 October 2005 (UTC)[reply]

EGT studies the dynamics and equilibria of games played by populations of players. The strategies players employ in the games determine their interdependent payoff or fitness. In contrast with the traditional applications of game theory, the players do not act rationally when choosing their strategies, but act instead according to a pre-programmed behavior pattern. A pure strategy is encoded in an individual's genome, which can evolve over time while repeatedly playing a game against other players in a population. The more successful strategies result in more offspring and the game is then iterated and studied.
Recently, evolutionary game theory has increasingly become of interest to economists, sociologists, and anthropologists, as well as philosophers. In these expanded contexts, the ‘evolution’ treated by EGT can refer to systems such as cultural evolution or economic behavior.

I also changed evolutionarily stable state to evolutionarily stable strategy, the former may have been correct. Given the lack of lucid discussion (errrr, any discussion really) on the distinction page for ESS, I used the term that is well defined, rather than the one no one has provided a literature reference or definition for... Pete.Hurd 04:33, 1 October 2005 (UTC)[reply]

Ummm, ok, state was correct, changed back, really ought to fix ESS page... Pete.Hurd 04:39, 1 October 2005 (UTC)[reply]

Economists, biologists, dynamics, and equilibria

[edit]
Thank you for your interest in this article, I'm glad to see someone is coming by our little corner! With respect to the claim that more economists use it than biologists: we had a brief discussion at Talk:Game theory about it and the referenced google scholar search there is helpful. Besides that, User:Pete.Hurd is a real live biologist working in game theory and he added the comment to the article.
With respect to the focus question, this goes more to motivation than to content I think. The idea in traditional game theory is that one uses equilibrium concepts to solve a game and then says "people will play equilibrium strategies." In my opinion, this last statement is incredibly unsatisfying since it doesn't do much to explain why people will play equilibrium strategies. Combined with the fact that many people don't appear to play equilibrium strategies, this naturally leads one to ask: what causes people to play the strategies that they do? A dynamic answer is one of the available explanations. So, its right to say that the fact that x is a equilibrium is part of the explanation of why people are playing strategy x. But its equally part of the explanation to say that the reason people play x is because their dynamics of strategy change resulting in them converging to x.
With respect to your other comments, I invite you to improve the article in anyway you see fit. We are very overwhelmed with work at Wikipedia:WikiProject Game theory, but you could also add a note there. (Despite the list of members there are really only two of us contributing on a regular basis to game theory articles.) Since you seem to know something about the area, perhaps you would like to join and help out?
:) --best, kevin [kzollman][talk] 17:53, 4 December 2005 (UTC)[reply]
Hi.
1) EGT used more by economists than biologists: I havn't collected quantitative data, but it's my strong impression (fallible, and subject to confirmation biases) that those in working in econ departments use EGT far more than those in biology departments, that a very small proportion of articles using EGT are produced by biologists. I could be wrong. My motivation when putting that sentence in was (as I recall) that many of the game theory pages asserted, or strongly implied that EGT was game theory applied to biological applications, which is obviously not true. I may have over-compensated (but I don't think so).
2) Focus on dynamics rather than equilibria: You have a point here. Something about differences in emphasis on dynamics, relative sizes of basins of attraction etc between EGT and traditional game theory might be a better comparision than between dynamics and equilibria in EGT. You probably know better than I do...
3) Lots of references, little content: Sure is, the article ought to have a stub tag on it. The references include the early and inflential articles (Taylor & Maynard Smith) and more recent authoritative texts, and represent more what would be appropriate to cite in a complete article than what is necessary to support what is curently written.
It would be nice if you could spare some time fixing this. It's outside my area of core competence, and not high on my list of things to learn. Pete.Hurd 21:19, 4 December 2005 (UTC)[reply]
The references list may have been intended as 'further reading,' which may have value for inclusion.--Nectar 01:30, 5 December 2005 (UTC)[reply]

Formulas are not properly annotated

[edit]

I see a nice formula p(x)=e^(-x/V)/V and keep wondering what "x" stands for. Formulas should be annotated in situ. Hunting the whole text for an explanation is a waste of time!— Preceding unsigned comment added by MilanCela (talkcontribs) 12:29, 27 September 2018 (UTC)[reply]