GENIUS FORUMS

Diebert,

But I believe there is no situation, by definition, that is so perfectly symmetric as the paradox suggests. The circumstances dictate the rational course of action and will as phenomenas always be unbalanced, leaning toward a certain 'way', not necessarily the best way of course, as David was saying: rationality doesn't equal omni-science.

It's not hard to think of symmetric situations, like choosing between two doors on a game show or something.

Nooooo, not the Monty Hall Problem!

Tricky!

Matt Gregory wrote:Diebert,

But I believe there is no situation, by definition, that is so perfectly symmetric as the paradox suggests. The circumstances dictate the rational course of action and will as phenomenas always be unbalanced, leaning toward a certain 'way', not necessarily the best way of course, as David was saying: rationality doesn't equal omni-science.

It's not hard to think of symmetric situations, like choosing between two doors on a game show or something.

I'd rather have a more realistic situation. Is it coincidence that you come up with something virtual and mental like a game show that's a very artificial, sterile and contrived situation? Like putting two dots on a piece of paper and claim symmetry!

the funny thing about the Monty Hallo Problem is that you see how quickly the symmetry is lost when things are getting added, or in a general sense information is leaked. I hate how this particular problem goes against intuition though!

Diebert van Rhijn wrote:I'd rather have a more realistic situation. Is it coincidence that you come up with something virtual and mental like a game show that's a very artificial, sterile and contrived situation? Like putting two dots on a piece of paper and claim symmetry!

Well, actually this is the sort of thing that comes up in business all the time. You have two options, two plans, or two investment opportunities; but you have enough money for one of them, so you must choose, hoping to maximize your return.

One plan is probably better than the other, but it is dependant on future conditions that can only be estimated. Your estimates are not significantly different. Or you can make projections, but making accurate projections is very costly (relative to the difference between the values of the projects).

If the expected return (from options A and B) is greater than zero, then you have a higher expected return from choosing one (at random, or however you like to make a choice - maybe ask Ganesh for guidance) rather than not making any investment, which you know will have zero return.

In fact, this is a general principle of all kinds of leadership - it is just better to make a choice and stick with it, when the cost of information is higher than the difference in outcome gained by that information. The rational thing to do in such a situation is to make a decision, because you have possibilities that you expect to have better outcomes than not making a decision, even if you don't know it that particular decision is "optimal" in the sense of being known to be better than the other possibilities you are aware of.


For concrete examples, you could be a bank deciding between two applicants for a large loan, or an insurance company deciding between covering two large projects or buildings, or an investor choosing between two possible large construction projects. You could be a person buying a house to live in, choosing between several properties.

Decisions normally have to be made with limited information. But there is a saying that decisions made with limited (partial) information are better than decisions made with no information at all.


Anyway, if you starve when you don't have to, you're an ass.

Diebert van Rhijn wrote:

Matt Gregory wrote:Diebert,

But I believe there is no situation, by definition, that is so perfectly symmetric as the paradox suggests. The circumstances dictate the rational course of action and will as phenomenas always be unbalanced, leaning toward a certain 'way', not necessarily the best way of course, as David was saying: rationality doesn't equal omni-science.

It's not hard to think of symmetric situations, like choosing between two doors on a game show or something.

I'd rather have a more realistic situation. Is it coincidence that you come up with something virtual and mental like a game show that's a very artificial, sterile and contrived situation? Like putting two dots on a piece of paper and claim symmetry!

All that's required of the choices is enough ignorance of them to make them practically the same. Let's say you're completely lost and you come to a fork in the road. Which way do you go? Is it rational to choose one direction over another? Say you want to find a particular book and there are two libraries near your house. Which library do you try first? Is it rational to choose one over the other? There are endless possibilities.

the funny thing about the Monty Hallo Problem is that you see how quickly the symmetry is lost when things are getting added, or in a general sense information is leaked. I hate how this particular problem goes against intuition though!

That's a really interesting problem, and I can see your point, but I don't think it matters in a situation where we're just trying to prove to someone that rationality isn't limited by the point of view in which the problem is expressed.

DHodges wrote: One plan is probably better than the other, but it is dependant on future conditions that can only be estimated. Your estimates are not significantly different. Or you can make projections, but making accurate projections is very costly (relative to the difference between the values of the projects).

An estimation is not the same as a pure 50/50 gamble. Most if not every real-life situation (so not an abstract simulation like a poker game, etc) that comes up, carrying any significance, seems unbalanced, no matter how tiny and subtle the way, no matter how it's related to point of view or estimations. Actually the points of view brought in by participants already can create a huge asymmetrical factor. For example: someone might be blind at one eye.

If the expected return (from options A and B) is greater than zero, then you have a higher expected return from choosing one (at random, or however you like to make a choice - maybe ask Ganesh for guidance) rather than not making any investment, which you know will have zero return.

Yes, so there's a reason to choose between AB and nothing at all. There's no symmetry and people will go with AB. The choice between A and B has again a different dynamic, and the more realistic an example would be the more you would notice how asymmetrical they appear, which doesn't mean to say everyone would make the same choice.

For concrete examples, you could be a bank deciding between two applicants for a large loan, or an insurance company deciding between covering two large projects or buildings, or an investor choosing between two possible large construction projects. You could be a person buying a house to live in, choosing between several properties.

In every case, unless there's an absurd lack of information, one choice will seem better or safer or more interesting than the other, according to what someone values and if those values are well developed. If one doesn't know what is wanted or valued, indecision might appear between two seemingly equal options. I think that might be related to the lack of consciousness of the situation and therefore a lack of possible rationality.

Decisions normally have to be made with limited information. But there is a saying that decisions made with limited (partial) information are better than decisions made with no information at all.

That limited information, in any realistic amounts, will cause perceived differences between the options and always provides some reason to go with one and not the other. Again, depending on how far the choice matters at all and the concreteness of the situation.

Matt Gregory wrote: All that's required of the choices is enough ignorance of them to make them practically the same.

That level of ignorance is what I'd suspect to cause a lack of reality or relevance. Like a poker game, which is a contrived simulation, no matter how high the stakes. Information is hidden on purpose and with agreement of participants. Even at poker of course at some point many personal factors play a role, reading body language and behavior and so on. This is why it's popular and computers suck at it. The participants break the symmetry the theory has.

Let's say you're completely lost and you come to a fork in the road. Which way do you go? Is it rational to choose one direction over another? Say you want to find a particular book and there are two libraries near your house. Which library do you try first? Is it rational to choose one over the other? There are endless possibilities.

The details in any of those examples will break symmetry. The libraries will not exactly be at the same distance with the exact same books with the exact same robots working there. One might even be in a safer neighborhood, etc. Same with a fork in the road, there's always a possible clue since the roads won't look exactly the same, some detail in the sky, etc.

It doesn't matter if the right road or library is taken, what matters if there's enough detail to use for a rational decision when one is required. And at the moment I just see no real-life (always multi-layered and subjective) example that can be stripped from the always present details that assist in making some kind of decision, at least the moment we care about taking one consciously.