9 ms·

How to choose?

120 points by fossuser 12y ago

tbrownaw 12y ago
If you know more than nothing, the best choice is often still random -- just with different weights applied to the various options, which I hear can be calculated with something called game theory.
- afafsd 12y ago
  Under what circumstances is that the case? Let's suppose, for instance, that two roads diverge in a yellow wood and you have reliable information that road A is the correct path with 90% probability and road B is the correct path with 10% probability. The normal person takes road A and will be right 90% of the time. On the other hand, if you try to be probabilistic about it and follow road B with 10% probability to reflect its 10% probability of being the right road, then your chances of picking the right road are 0.90.9 + 0.10.1 = 82%. Maybe there's a more complex situation where the weighted random decision making is a better idea, but I'm not seeing it right now.
  - Sambdala 12y ago
    Since my background is poker, I'll use an (overly simplified) example from there. There are situations where you want to manipulate your range (set of hands) in a given spot so that your actions make up a certain distribution. For example, if you want your range to be made up of 30% bluffs, you can figure out how often you get to that spot with a bluff and how often you get to that spot with a strong hand. Since you're always continuing with the strong hand, you can then know what percentage of the time you need to continue to bluff in order to accomplish this. Since it doesn't matter which of those hands you choose to bluff with, it generally makes sense to use randomness to decide in the moment.
    - TeMPOraL 12y ago
      I don't play poker, but from what I understand, the reason for including randomness in your decision is to throw your opponents off. I.e. You're playing against people who try to predict your behavior, so you throw in a random component to make it harder for them. In decisions not involving active adversaries, deciding at random doesn't give you anything over a deterministic solution.
      - antimagic 12y ago
        Nearly, but just to nitpick, you're not including randomness, you're actively making a choice to make the resulting pattern difficult to distinguish from randomness. So, sometimes you raise by a large amount because you have a strong hand. You don't want other players to be identify that you have a strong hand by the fact that you just offered a large raise, so you have to also bluff sometimes, raising when you really don't have the cards. Your actions are not random, but they are perceived as random by the other players.
  - lmm 12y ago
    It happens when you're competing. Suppose you're trying to find some buried treasure that you know at the end of one of the roads. But another tribe is trying to kill you, and will be waiting at the end of one of the roads. In that case your best strategy (and theirs) is to make the choice of which road to take probabilistically.
    - TeMPOraL 12y ago
      Well, this is a much complicated problem. You need to answer at least following questions: - what is your utility of finding a treasure and not getting killed? - what is your utility of not finding a treasure and not getting killed? - does the other tribe know your utility distributions? I'm pretty sure there's a solution to this that involves probability and fixed points but does not involve flipping a coin to chose a road.
dkarapetyan 12y ago
Anyone see an analogy here with startups? I do.
- camillomiller 12y ago
  The difference is that investors would like us to believe they follow a rational path to decide where to invest :D
  - mareofnight 12y ago
    Or in the case of VC firms, would like their investors to believe this.
- dredmorbius 12y ago
  It's one of the reasons success is such a miserable teacher. Much of success is random.
tikhonj 12y ago
I regularly use randomness to make (small) decisions in my life. I look at my watch and decide based on the last digit of the seconds. (Often, I just want one bit—even or odd.) Without this, I'm indecisive all too often, in the stupidest of places. Choosing what to order at a restaurant, for example, used to be a bit of a pain; now I just select a few acceptable choices and then decide randomly. Similarly, it's also how I vary up my walk home: even, I take a left turn, odd—a right. It's a surprisingly useful little trick!
- lotharbot 12y ago
  My parents used the last digit on a stopwatch to decide which of the kids got to pick the radio station in the car and similarly trivial decisions. (If 0-9 couldn't be divided evenly among the kids present, leftover digits would result in a re-roll.) This trick takes away two types of pressure: (1) trying to remember past decisions to be fair. Over a long enough period of time, with a fair coin/die/watch, everyone will get about the same chances. Or you'll walk all the routes about equally. Or you'll equally sample all of the acceptable menu items. (2) trying to justify which decision was made. There's no need for explaining why Bob got to sit in the front seat, or why you ordered the chicken, or why you walked on 3rd street instead of 2nd. There's no need for second-guessing whether your decision accounted for all of the information and was really reasoned through correctly. I saw the true value of removing the "justify every decision" step when a friend of mine was suffering from a severe mental illness that made it difficult to make trivial decisions. The result of trying to justify every decision: standing in the kitchen for several minutes because "I want to make a peanut butter sandwich, but I can't decide whether to get the bread or the knife first." By creating an automatic process for making decisions that are only slightly less trivial, we free ourselves to focus on the more important things in life.
  - lutusp 12y ago
    > My parents used the last digit on a stopwatch to decide which of the kids got to pick the radio station in the car and similarly trivial decisions. (If 0-9 couldn't be divided evenly among the kids present, leftover digits would result in a re-roll.) Easily solved by taking the [EDIT: seconds digits] modulo the number of children, but I guess that might be difficult while driving. EDIT: People, please try to think this through before posting an innumerate reply.
vecter 12y ago
This seems relevant: In the days when Sussman was a novice, Minsky once came to him as he sat hacking at the PDP-6. "What are you doing?", asked Minsky. "I am training a randomly wired neural net to play Tic-tac-toe", Sussman replied. "Why is the net wired randomly?", asked Minsky. "I do not want it to have any preconceptions of how to play", Sussman said. Minsky then shut his eyes. "Why do you close your eyes?" Sussman asked his teacher. "So that the room will be empty." At that moment, Sussman was enlightened. [0] http://en.wikipedia.org/wiki/Hacker_koan http://en.wikipedia.org/wiki/Hacker_koan
- blinduck 12y ago
  I don't quite understand this. A person's preconceptions is the collection of his previous knowledge about the subject. What does it mean to have random preconceptions?
  - grey-area 12y ago
    I think the point is that you can't remove external constraints from a situation simply by willing it (make the people in the room disappear, make your AI independent of assumptions). For example the type of neural net chosen, the reinforcement and other parameters, the rules of the game chosen by the creator are all constraints which are not randomly chosen, so randomly choosing the starting state is not necessarily going to remove preconceptions built in to the model, and it is still choosing some starting state, just not an arbitrary one chosen by the creator. I'm not sure it really applies to this article though, which is about a different problem - saving the time spent worrying about unimportant decisions or those where you don't have enough inputs and choosing to take a random path.
  - Terr_ 12y ago
    I think it means that: (A) There are always "preconceptions" present, even if the programmer isn't working on a level of abstraction that lets them consciously realize it. (B) Programming a system to be too naive can backfire.
  - sanityinc 12y ago
RK 12y ago
My advisor in grad school once gave the following advice algorithm for making a decision: 1. Flip a coin. 2. If the result of the coin flip makes you hesitate at all, you know that was the choice you didn't really like anyway. Go with the other choice.
- baddox 12y ago
  The problem is that in any decision difficult enough to make you resort to a coin flip, most likely either coin flip result will make you hesitate.
  - duncanawoods 12y ago
    I agree but I wouldn't take it too literally. The coin flip isn't binding but it can be fascinatingly revealing about your own opinions. I see it as a way to actualise the consequences of the choice and cut through layers of intellectual abstraction. In some respects its a tool to let you engage emotional thinking to help make better decisions. On the flip, it can suddenly trigger a feeling of loss and regret. Our fundamental beliefs can be strangely out of reach when we think too hard. A similar tool is just explaining your decision to another. I can get a flush of emotion e.g. embarrassment or shame, that you don't get when you just cogitate alone. Pretty useful for tough design decisions e.g. midway through explaining a particularly clever idea I find myself apologising... its time to rethink things!
  - arcatek 12y ago
    Actually, the real point of flipping a coin is that during the brief lapse of time during which the piece is in the air, you will know what's the choice you prefer. The result doesn't matter.
  - TeMPOraL 12y ago
    Also if both decisions make you hesitate, it suggests that costs/benefits of both of them cancel each other out, so you may as well stop hesitating and go with what the coin tells you.
- hawkice 12y ago
kissickas 12y ago
The author never comes back to it, but the Greeks used a lottery to choose some of their politicians in order to mitigate corruption: https://en.wikipedia.org/wiki/Athenian_democracy#Selection_by_lot_.28allotment.29 https://en.wikipedia.org/wiki/Athenian_democracy#Selection_b... I discovered this the other day when trying to find out what "drawing lots" actually meant in the Bible, another form of letting chance decide things. Apparently we don't know exactly what people were using back then.
- blue1 12y ago
  The Republic of Venice developed that idea to a system of great complexity for electing the Doge, that included copious injections of randomness. There is an interesting analysis of this protocol in a paper by HP laboratories, available here (pdf): http://www.hpl.hp.com/techreports/2007/HPL-2007-28R1.pdf http://www.hpl.hp.com/techreports/2007/HPL-2007-28R1.pdf
shire 12y ago
I think I'm very indecisive, I have hard time ordering stuff at restaurants without holding up lines. I need to find a method of deciding soon. I'll try the coin trick.
- ownagefool 12y ago
  Having someone at the front of a queue dithering is probably one of the most "much more annoying than it should be things". ;p
- gglitch 12y ago
  I have that problem too. I usually narrow it down to two or three options and ask the server/cashier what I want. About 80% of the time I go with whatever s/he says; other times, simply hearing his/her recommendation makes me realize what my actual preference is.
dredmorbius 12y ago
The fairness and justification elements alone are fascinating. This also brings to mind strongly Suzanne Vega's "Predictions."
Gmo 12y ago
Makes me think of this book (which turns out pretty badly) : http://en.wikipedia.org/wiki/The_Dice_Man http://en.wikipedia.org/wiki/The_Dice_Man
- slazaro 12y ago
  I recommend this book. It's really entertaining and thought provoking. I remember carrying a die everywhere a few weeks after reading this book, in order to add some randomness to my decisions. Silly, but fun!
dreamfactory2 12y ago
Isn't this covered by Cynefin - http://en.wikipedia.org/wiki/Cynefin http://en.wikipedia.org/wiki/Cynefin? Action first in the chaotic quadrant corresponds to making a fast decision in the face of unpredictability.
- pontifier 12y ago
  I love finding new ways of thinking and solving problems. Thanks for referencing this.
  - dreamfactory2 12y ago
    If you are in software dev worth reading http://lizkeogh.com/2012/03/11/cynefin-for-devs/ http://lizkeogh.com/2012/03/11/cynefin-for-devs/

120 points by fossuser 12y ago

If you know more than nothing, the best choice is often still random -- just with different weights applied to the various options, which I hear can be calculated with something called game theory.

afafsd 12y ago
Under what circumstances is that the case? Let's suppose, for instance, that two roads diverge in a yellow wood and you have reliable information that road A is the correct path with 90% probability and road B is the correct path with 10% probability. The normal person takes road A and will be right 90% of the time. On the other hand, if you try to be probabilistic about it and follow road B with 10% probability to reflect its 10% probability of being the right road, then your chances of picking the right road are 0.90.9 + 0.10.1 = 82%. Maybe there's a more complex situation where the weighted random decision making is a better idea, but I'm not seeing it right now.
- Sambdala 12y ago
  Since my background is poker, I'll use an (overly simplified) example from there. There are situations where you want to manipulate your range (set of hands) in a given spot so that your actions make up a certain distribution. For example, if you want your range to be made up of 30% bluffs, you can figure out how often you get to that spot with a bluff and how often you get to that spot with a strong hand. Since you're always continuing with the strong hand, you can then know what percentage of the time you need to continue to bluff in order to accomplish this. Since it doesn't matter which of those hands you choose to bluff with, it generally makes sense to use randomness to decide in the moment.
  - TeMPOraL 12y ago
    I don't play poker, but from what I understand, the reason for including randomness in your decision is to throw your opponents off. I.e. You're playing against people who try to predict your behavior, so you throw in a random component to make it harder for them. In decisions not involving active adversaries, deciding at random doesn't give you anything over a deterministic solution.
    - antimagic 12y ago
      Nearly, but just to nitpick, you're not including randomness, you're actively making a choice to make the resulting pattern difficult to distinguish from randomness. So, sometimes you raise by a large amount because you have a strong hand. You don't want other players to be identify that you have a strong hand by the fact that you just offered a large raise, so you have to also bluff sometimes, raising when you really don't have the cards. Your actions are not random, but they are perceived as random by the other players.
- lmm 12y ago
  It happens when you're competing. Suppose you're trying to find some buried treasure that you know at the end of one of the roads. But another tribe is trying to kill you, and will be waiting at the end of one of the roads. In that case your best strategy (and theirs) is to make the choice of which road to take probabilistically.
  - TeMPOraL 12y ago
    Well, this is a much complicated problem. You need to answer at least following questions: - what is your utility of finding a treasure and not getting killed? - what is your utility of not finding a treasure and not getting killed? - does the other tribe know your utility distributions? I'm pretty sure there's a solution to this that involves probability and fixed points but does not involve flipping a coin to chose a road.

dkarapetyan 12y ago

Anyone see an analogy here with startups? I do.

camillomiller 12y ago
The difference is that investors would like us to believe they follow a rational path to decide where to invest :D
- mareofnight 12y ago
  Or in the case of VC firms, would like their investors to believe this.
dredmorbius 12y ago
It's one of the reasons success is such a miserable teacher. Much of success is random.

tikhonj 12y ago

I regularly use randomness to make (small) decisions in my life. I look at my watch and decide based on the last digit of the seconds. (Often, I just want one bit—even or odd.) Without this, I'm indecisive all too often, in the stupidest of places. Choosing what to order at a restaurant, for example, used to be a bit of a pain; now I just select a few acceptable choices and then decide randomly. Similarly, it's also how I vary up my walk home: even, I take a left turn, odd—a right. It's a surprisingly useful little trick!

lotharbot 12y ago
My parents used the last digit on a stopwatch to decide which of the kids got to pick the radio station in the car and similarly trivial decisions. (If 0-9 couldn't be divided evenly among the kids present, leftover digits would result in a re-roll.) This trick takes away two types of pressure: (1) trying to remember past decisions to be fair. Over a long enough period of time, with a fair coin/die/watch, everyone will get about the same chances. Or you'll walk all the routes about equally. Or you'll equally sample all of the acceptable menu items. (2) trying to justify which decision was made. There's no need for explaining why Bob got to sit in the front seat, or why you ordered the chicken, or why you walked on 3rd street instead of 2nd. There's no need for second-guessing whether your decision accounted for all of the information and was really reasoned through correctly. I saw the true value of removing the "justify every decision" step when a friend of mine was suffering from a severe mental illness that made it difficult to make trivial decisions. The result of trying to justify every decision: standing in the kitchen for several minutes because "I want to make a peanut butter sandwich, but I can't decide whether to get the bread or the knife first." By creating an automatic process for making decisions that are only slightly less trivial, we free ourselves to focus on the more important things in life.
- lutusp 12y ago
  > My parents used the last digit on a stopwatch to decide which of the kids got to pick the radio station in the car and similarly trivial decisions. (If 0-9 couldn't be divided evenly among the kids present, leftover digits would result in a re-roll.) Easily solved by taking the [EDIT: seconds digits] modulo the number of children, but I guess that might be difficult while driving. EDIT: People, please try to think this through before posting an innumerate reply.

vecter 12y ago

This seems relevant: In the days when Sussman was a novice, Minsky once came to him as he sat hacking at the PDP-6. "What are you doing?", asked Minsky. "I am training a randomly wired neural net to play Tic-tac-toe", Sussman replied. "Why is the net wired randomly?", asked Minsky. "I do not want it to have any preconceptions of how to play", Sussman said. Minsky then shut his eyes. "Why do you close your eyes?" Sussman asked his teacher. "So that the room will be empty." At that moment, Sussman was enlightened. [0] http://en.wikipedia.org/wiki/Hacker_koan http://en.wikipedia.org/wiki/Hacker_koan

blinduck 12y ago
I don't quite understand this. A person's preconceptions is the collection of his previous knowledge about the subject. What does it mean to have random preconceptions?
- grey-area 12y ago
  I think the point is that you can't remove external constraints from a situation simply by willing it (make the people in the room disappear, make your AI independent of assumptions). For example the type of neural net chosen, the reinforcement and other parameters, the rules of the game chosen by the creator are all constraints which are not randomly chosen, so randomly choosing the starting state is not necessarily going to remove preconceptions built in to the model, and it is still choosing some starting state, just not an arbitrary one chosen by the creator. I'm not sure it really applies to this article though, which is about a different problem - saving the time spent worrying about unimportant decisions or those where you don't have enough inputs and choosing to take a random path.
- Terr_ 12y ago
  I think it means that: (A) There are always "preconceptions" present, even if the programmer isn't working on a level of abstraction that lets them consciously realize it. (B) Programming a system to be too naive can backfire.
- sanityinc 12y ago

RK 12y ago

My advisor in grad school once gave the following advice algorithm for making a decision: 1. Flip a coin. 2. If the result of the coin flip makes you hesitate at all, you know that was the choice you didn't really like anyway. Go with the other choice.

baddox 12y ago
The problem is that in any decision difficult enough to make you resort to a coin flip, most likely either coin flip result will make you hesitate.
- duncanawoods 12y ago
  I agree but I wouldn't take it too literally. The coin flip isn't binding but it can be fascinatingly revealing about your own opinions. I see it as a way to actualise the consequences of the choice and cut through layers of intellectual abstraction. In some respects its a tool to let you engage emotional thinking to help make better decisions. On the flip, it can suddenly trigger a feeling of loss and regret. Our fundamental beliefs can be strangely out of reach when we think too hard. A similar tool is just explaining your decision to another. I can get a flush of emotion e.g. embarrassment or shame, that you don't get when you just cogitate alone. Pretty useful for tough design decisions e.g. midway through explaining a particularly clever idea I find myself apologising... its time to rethink things!
- arcatek 12y ago
  Actually, the real point of flipping a coin is that during the brief lapse of time during which the piece is in the air, you will know what's the choice you prefer. The result doesn't matter.
- TeMPOraL 12y ago
  Also if both decisions make you hesitate, it suggests that costs/benefits of both of them cancel each other out, so you may as well stop hesitating and go with what the coin tells you.
hawkice 12y ago

kissickas 12y ago

The author never comes back to it, but the Greeks used a lottery to choose some of their politicians in order to mitigate corruption: https://en.wikipedia.org/wiki/Athenian_democracy#Selection_by_lot_.28allotment.29 https://en.wikipedia.org/wiki/Athenian_democracy#Selection_b... I discovered this the other day when trying to find out what "drawing lots" actually meant in the Bible, another form of letting chance decide things. Apparently we don't know exactly what people were using back then.

blue1 12y ago
The Republic of Venice developed that idea to a system of great complexity for electing the Doge, that included copious injections of randomness. There is an interesting analysis of this protocol in a paper by HP laboratories, available here (pdf): http://www.hpl.hp.com/techreports/2007/HPL-2007-28R1.pdf http://www.hpl.hp.com/techreports/2007/HPL-2007-28R1.pdf

shire 12y ago

I think I'm very indecisive, I have hard time ordering stuff at restaurants without holding up lines. I need to find a method of deciding soon. I'll try the coin trick.

ownagefool 12y ago
Having someone at the front of a queue dithering is probably one of the most "much more annoying than it should be things". ;p
gglitch 12y ago
I have that problem too. I usually narrow it down to two or three options and ask the server/cashier what I want. About 80% of the time I go with whatever s/he says; other times, simply hearing his/her recommendation makes me realize what my actual preference is.

dredmorbius 12y ago

The fairness and justification elements alone are fascinating. This also brings to mind strongly Suzanne Vega's "Predictions."

Gmo 12y ago

Makes me think of this book (which turns out pretty badly) : http://en.wikipedia.org/wiki/The_Dice_Man http://en.wikipedia.org/wiki/The_Dice_Man

slazaro 12y ago
I recommend this book. It's really entertaining and thought provoking. I remember carrying a die everywhere a few weeks after reading this book, in order to add some randomness to my decisions. Silly, but fun!

dreamfactory2 12y ago

Isn't this covered by Cynefin - http://en.wikipedia.org/wiki/Cynefin http://en.wikipedia.org/wiki/Cynefin? Action first in the chaotic quadrant corresponds to making a fast decision in the face of unpredictability.

pontifier 12y ago
I love finding new ways of thinking and solving problems. Thanks for referencing this.
- dreamfactory2 12y ago
  If you are in software dev worth reading http://lizkeogh.com/2012/03/11/cynefin-for-devs/ http://lizkeogh.com/2012/03/11/cynefin-for-devs/