Re: evolution was: [ProgSoc] Video yum-cha

[alister air <baitoven@nospam.progsoc.uts.edu.au>]

On Tue, 9 Apr 2002, Justin Warren wrote:

> > This highlights one of the obvious flaws in rational decision-making (and
> > not just as it applies to games).  
> 
> Prisoner's Dilemma is really about short term decision making where the 
> actors have no expectations about future interactions or collaborations, 
> not long term affects.

My point about the rational process stands even in a once-off
relationship.  The one-off relationship is not, I would think,
zero-sum.  It could be, but the issues stand outside of whether it's a
zero-sum game.  They're just more obvious when total jail time served is
less in years if both players keep silent.  If both keep silent and get
two years, that's better than if one defects - total time is ten years
(just for the other person) or if both defect and get three years (total
time = six years).  

> The iterative Prisoner's Dilemma studies a different issue, that of
> blind variation and selective retention evolution to examine possible
> methods in such an environment. The difference between classic and
> iterative Prisoner's Dilemma takes you into the realm of the principle
> of suboptimisation and 'the whole is more than the sum of its parts'.
>
> Classic Prisoner's Dilemma is usually formulated as a zero-sum game,
> whereas the iterative model is non-zero-sum.

But did you know the technique that beat Tit-for-Tat?  For the
rationalists amongst you, it's counterintuitive.

Alister

--

Welfare doesn't work, because it gives | Illiteracy blinds the spirit. 
the poor incentives to remain poor.    | Study, soldier. (CNT leaflet)
Instead, let's give the wealthy a huge |
tax cut, so the poor will have an      | It's never too late
incentive to become millionaires!      | To smash the state.



-
You are subscribed to the progsoc mailing list. To unsubscribe, send a
message containing "unsubscribe" to progsoc-request@nospam.progsoc.uts.edu.au.
If you are having trouble, ask owner-progsoc@nospam.progsoc.uts.edu.au for help.