Welcome Guest [Log In] [Register]
Welcome to The New Coffee Room. We hope you enjoy your visit.


You're currently viewing our forum as a guest. This means you are limited to certain areas of the board and there are some features you can't use. If you join our community, you'll be able to access member-only sections, and use many member-only features such as customizing your profile, sending personal messages, and voting in polls. Registration is simple, fast, and completely free.


Join our community!


If you're already a member please log in to your account to access all of our features:

Username:   Password:
Add Reply
Prisoner's Dilemma
Topic Started: Apr 12 2010, 09:04 AM (320 Views)
John Galt
Fulla-Carp
I've never heard of the game called Prisoner's Dilemma that was mentioned in this article: http://bigthink.com/ideas/19541

Sounds intriguing; I'd love to learn more.
Let us begin anew, remembering on both sides that civility is not a sign of weakness.
Offline Profile Quote Post Goto Top
 
Klaus
Member Avatar
HOLY CARP!!!
It's a classic from game theory.

I think the prisoner's dilemma illustrates that capitalism (in the sense of: individuals trying to maximize their profits) will sometimes produce suboptimal results.
Trifonov Fleisher Klaus Sokolov Zimmerman
Offline Profile Quote Post Goto Top
 
Piano*Dad
Member Avatar
Bull-Carp
It doesn't really have anything to do with capitalism. You can discover prisoners' dilemmas all over the place in just about every circumstance imaginable.
Offline Profile Quote Post Goto Top
 
John Galt
Fulla-Carp
Any good books to recommend about game theory in general? For the layperson?
Let us begin anew, remembering on both sides that civility is not a sign of weakness.
Offline Profile Quote Post Goto Top
 
jon-nyc
Member Avatar
Cheers
I thin its most fascinating application is evolutionary biology. Particularly the concept of the 'evolutionarily stable strategy'.
In my defense, I was left unsupervised.
Online Profile Quote Post Goto Top
 
Red Rice
HOLY CARP!!!
Here's a good article by Carl Sagan, though it addresses the social strategy of game theory more than the mathematics. Copy and paste:

www.dallasuu.org/re/adult/huumanist/TitTatSagan.pdf
Civilisation, I vaguely realized then - and subsequent observation has confirmed the view - could not progress that way. It must have a greater guiding principle to survive. To treat it as a carcase off which each man tears as much as he can for himself, is to stand convicted a brute, fit for nothing better than a jungle existence, which is a death-struggle, leading nowhither. I did not believe that was the human destiny, for Man individually was sane and reasonable, only collectively a fool.

I hope the gunner of that Hun two-seater shot him clean, bullet to heart, and that his plane, on fire, fell like a meteor through the sky he loved. Since he had to end, I hope he ended so. But, oh, the waste! The loss!

- Cecil Lewis
Offline Profile Quote Post Goto Top
 
Kincaid
Member Avatar
HOLY CARP!!!
This made me think of the TV show "Survivor". I've never missed a season because I find the human interactions so fascinating.

On one episode, after the two "tribes" were merged and they were all playing individually (but with certain alliances in place) they played for a reward of a vehicle. The person that won the car/truck was then given a choice. Give up the car/truck and all of the opponents would get the same vehicle for their own.

Setting aside for the moment the show's history - where anyone that won a vehicle was the next person voted out - and setting aside the idea that getting everyone a vehicle might get you farther in the game because people are more kindly disposed toward you because of your sacrifice (and your gain), I personally felt I would have instantly decided to give everyone the great, and easy gift of the vehicle.

I was surprised that the person kept the car and deprived the other six or so people of it. I think I would have given up the vehhicle without a second thought.

FWIW, that contestant did get voted out next, IIRC.
Edited by Kincaid, Apr 12 2010, 09:55 AM.
Kincaid - disgusted Republican Partisan since 2006.
Offline Profile Quote Post Goto Top
 
Klaus
Member Avatar
HOLY CARP!!!
Piano*Dad
Apr 12 2010, 09:09 AM
It doesn't really have anything to do with capitalism. You can discover prisoners' dilemmas all over the place in just about every circumstance imaginable.
In my point of view, one of the basic assumptions underlying capitalism is that the global well-being is maximised if every individual maximises his or her well-being.

The prisoner dilemma illustrates that this assumption is sometimes wrong and global planning would be superior in such circumstances.
Trifonov Fleisher Klaus Sokolov Zimmerman
Offline Profile Quote Post Goto Top
 
Red Rice
HOLY CARP!!!
Klaus
Apr 12 2010, 10:02 AM
The prisoner dilemma illustrates that this assumption is sometimes wrong and global planning would be superior in such circumstances.
But only because the prisoner's dilemma is an artifically restricted scenario that eliminates communication.
Civilisation, I vaguely realized then - and subsequent observation has confirmed the view - could not progress that way. It must have a greater guiding principle to survive. To treat it as a carcase off which each man tears as much as he can for himself, is to stand convicted a brute, fit for nothing better than a jungle existence, which is a death-struggle, leading nowhither. I did not believe that was the human destiny, for Man individually was sane and reasonable, only collectively a fool.

I hope the gunner of that Hun two-seater shot him clean, bullet to heart, and that his plane, on fire, fell like a meteor through the sky he loved. Since he had to end, I hope he ended so. But, oh, the waste! The loss!

- Cecil Lewis
Offline Profile Quote Post Goto Top
 
Klaus
Member Avatar
HOLY CARP!!!
John Galt
Apr 12 2010, 09:10 AM
Any good books to recommend about game theory in general? For the layperson?
I've read some books about game theory, both "for the layperson" and more scientific books.

Unfortunately, in my experience all the "layperson" books are junk. The true essence of game theory is in the mathematics (e.g. notions such as Nash equilibirum, Pareto frontier etc.), and this is not captured properly by the informal presentations.

This is a book I like, but it is not an easy read:

Posted Image
Trifonov Fleisher Klaus Sokolov Zimmerman
Offline Profile Quote Post Goto Top
 
Klaus
Member Avatar
HOLY CARP!!!
Red Rice
Apr 12 2010, 10:06 AM
Klaus
Apr 12 2010, 10:02 AM
The prisoner dilemma illustrates that this assumption is sometimes wrong and global planning would be superior in such circumstances.
But only because the prisoner's dilemma is an artifically restricted scenario that eliminates communication.
Well, take for example price formation in a market with only a few producers. From the point of view of these producers, it would be best to coordinate with all other producers to negotiate a (high) fixed price. However, if each individual maximises his profit, he'll sell for a lower price than the others because then he can sell more. If everybody does this, this means that globally all the sellers "loose".

Of course, this is a somewhat misleading example since in this case the capitalistic solution will be superior if you consider the buyers, too, but it illustrates that the scenario does occur in practice.

Another example would be to consider advertizing in a market where the overall demand is fixed or almost fixed and advertizing only changes the relative market share of the companies (such as cigarettes, or headache tablets).
Trifonov Fleisher Klaus Sokolov Zimmerman
Offline Profile Quote Post Goto Top
 
Red Rice
HOLY CARP!!!
Klaus
Apr 12 2010, 10:15 AM
Well, take for example price formation in a market with only a few producers. From the point of view of these producers, it would be best to coordinate with all other producers to negotiate a (high) fixed price. However, if each individual maximises his profit, he'll sell for a lower price than the others because then he can sell more. If everybody does this, this means that globally all the sellers "loose".

Of course, this is a somewhat misleading example since in this case the capitalistic solution will be superior if you consider the buyers, too, but it illustrates that the scenario does occur in practice.
I see your point, but we have to distinguish between "communcation" (or "information") and "cooperation". They are not necessarily synonymous in either the Prisoner's Dilemma or the open market. In the open market, producers do not operate in a vacuum; they know each other's prices, each other's qualities/flaws, and receive continuous feedback from the marketplace. With that information, they can adjust their products/prices/marketing, so that in the end the sellers don't all lose, but rather achieve a Nash equilibrium (of course, some will lose, if their information gathering and execution are flawed).

Now, if you deprived producers of all the information they need to achieve that equilibrium, or impose a system where the information can't be implemented (such as communism), then yes, they would be operating in a scenario comparable to the Prisoner's Dilemma. But that just doesn't happen in an open market.
Civilisation, I vaguely realized then - and subsequent observation has confirmed the view - could not progress that way. It must have a greater guiding principle to survive. To treat it as a carcase off which each man tears as much as he can for himself, is to stand convicted a brute, fit for nothing better than a jungle existence, which is a death-struggle, leading nowhither. I did not believe that was the human destiny, for Man individually was sane and reasonable, only collectively a fool.

I hope the gunner of that Hun two-seater shot him clean, bullet to heart, and that his plane, on fire, fell like a meteor through the sky he loved. Since he had to end, I hope he ended so. But, oh, the waste! The loss!

- Cecil Lewis
Offline Profile Quote Post Goto Top
 
Klaus
Member Avatar
HOLY CARP!!!
Red Rice
Apr 12 2010, 10:25 AM
Klaus
Apr 12 2010, 10:15 AM
Well, take for example price formation in a market with only a few producers. From the point of view of these producers, it would be best to coordinate with all other producers to negotiate a (high) fixed price. However, if each individual maximises his profit, he'll sell for a lower price than the others because then he can sell more. If everybody does this, this means that globally all the sellers "loose".

Of course, this is a somewhat misleading example since in this case the capitalistic solution will be superior if you consider the buyers, too, but it illustrates that the scenario does occur in practice.
I see your point, but we have to distinguish between "communcation" (or "information") and "cooperation". They are not necessarily synonymous in either the Prisoner's Dilemma or the open market. In the open market, producers do not operate in a vacuum; they know each other's prices, each other's qualities/flaws, and receive continuous feedback from the marketplace. With that information, they can adjust their products/prices/marketing, so that in the end the sellers don't all lose, but rather achieve a Nash equilibrium (of course, some will lose, if their information gathering and execution are flawed).

Now, if you deprived producers of all the information they need to achieve that equilibrium, or impose a system where the information can't be implemented (such as communism), then yes, they would be operating in a scenario comparable to the Prisoner's Dilemma. But that just doesn't happen in an open market.
Oh, game theory (and the prisoner dilemma example) is not at all ignorant of what you call "communication". In game theory, your "communication" would be modeled as reactions to the decisions of other agents, e.g., an adjusted price as a reaction to the price change of another party. Such reactions to events adds a time dimension. In game theory terminology this is called an extensive-form game.

The simple prisoner dilemma does deliberately not model this time dimension, because it is orthogonal to the point of this dilemma. Game theorists call this a "simultaneous game". There are variants of the prisoners dilemma that turn it into an extensive-form game, such as the iterated prisoner's dilemma, where the opponents "play" the game many times in succession and remember previous choices of their opponents.
Trifonov Fleisher Klaus Sokolov Zimmerman
Offline Profile Quote Post Goto Top
 
Piano*Dad
Member Avatar
Bull-Carp
Klaus
Apr 12 2010, 10:02 AM
Piano*Dad
Apr 12 2010, 09:09 AM
It doesn't really have anything to do with capitalism. You can discover prisoners' dilemmas all over the place in just about every circumstance imaginable.
In my point of view, one of the basic assumptions underlying capitalism is that the global well-being is maximised if every individual maximises his or her well-being.

The prisoner dilemma illustrates that this assumption is sometimes wrong and global planning would be superior in such circumstances.
In my view, and the view of most economists, this isn't an assumption at all. It's a well-argued result that flows from many other deeper axioms (of rational choice, for instance).

The fact that there exist circumstances in which strategic interaction leads to socially sub-optimal results is quite well known, and it is not a threat to 'market economics.' The notion that markets can lead to better outcomes than planning is not threatened by special examples in which it does not.

If you want to read about how planning itself leads to pernicious prisoners' dilemmas, try Mancur Olson's book Power and Prosperity. It contains perhaps the best explanation of the rise and fall of Soviet central planning.

BTW, the Myerson book is quite good. I have used his work in class.
Offline Profile Quote Post Goto Top
 
Klaus
Member Avatar
HOLY CARP!!!
P*D, I am not at all advocating central planning or saying that this is a "threat to 'market economics'". Even though central planning could theoretically produce superior results in some situations, it is in general hard or impossible to identify these situations and gather all information required to make the "perfect plan". One should apply central planning only in very well-defined, well-understood situations, where it is clear that the free market will produce undesirable results. But the default should certainly be the free market.

So you can withdraw your "Piano*Dad defends capitalism" canon ;)
Trifonov Fleisher Klaus Sokolov Zimmerman
Offline Profile Quote Post Goto Top
 
Klaus
Member Avatar
HOLY CARP!!!
Piano*Dad
Apr 12 2010, 01:31 PM
In my view, and the view of most economists, this isn't an assumption at all. It's a well-argued result that flows from many other deeper axioms (of rational choice, for instance).
Do you mind to provide a source for your "the view of most economists" claim?

In my mathematics/CS background, I see capitalism as being similar to a class of so-called greedy algorithms, where the basic idea is to make the locally optimal choice (which corresponds to individual profit optimization) in each situation. For some problems, the greedy algorithm will produce the globally optimal solution, for other problems it produces terrible solutions. It is in all cases an algorithm which scales very well, meaning it works well for big problems - in contrast to algorithsm that need "global knowledge" (which is similar to the central planning problem).
Trifonov Fleisher Klaus Sokolov Zimmerman
Offline Profile Quote Post Goto Top
 
Piano*Dad
Member Avatar
Bull-Carp
Quote:
 
Do you mind to provide a source for your "the view of most economists" claim?


Yeah, any microeconomics textbook from principles up to Andrue Mas-Collel's massive graduate micro text. :whome:

The notion of the efficiency of a market equilibrium is no more 'assumed' than is the inefficiency of an arms race or any other prisoners' dilemma example of cooperation failure. These are results that flow from other deeper assumptions. In a sense, all theoretical models' outcomes are derived from a series of assumptions, but that doesn't make the outcomes simply assumed. They are the deductive results of the deeper axioms (like completeness and transitivity, for instance) coupled with an apparatus that presumes that individuals attempt to maximize something subject to constraints.
Offline Profile Quote Post Goto Top
 
brenda
Member Avatar
..............
Klaus, now that you have two children, you will have opportunities in the future for some home applications of the prisoner's dilemma. :)
“Weeds are flowers, too, once you get to know them.”
~A.A. Milne
Offline Profile Quote Post Goto Top
 
Piano*Dad
Member Avatar
Bull-Carp
brenda
Apr 12 2010, 04:53 PM
Klaus, now that you have two children, you will have opportunities in the future for some home applications of the prisoner's dilemma. :)
:devilgrin:
Offline Profile Quote Post Goto Top
 
brenda
Member Avatar
..............
Piano*Dad
Apr 12 2010, 05:32 PM
brenda
Apr 12 2010, 04:53 PM
Klaus, now that you have two children, you will have opportunities in the future for some home applications of the prisoner's dilemma. :)
:devilgrin:
I thought you might like that, PD. :lol2:
“Weeds are flowers, too, once you get to know them.”
~A.A. Milne
Offline Profile Quote Post Goto Top
 
« Previous Topic · The New Coffee Room · Next Topic »
Add Reply