Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

We think this cool study we found is flawed. Help us reproduce it

pudding.cool

Post reply on HN
Print view View original

171–180 of 361 posts

Re: We think this cool study we found is flawed. Help us reproduce it

#171

Post by elif »

prolyx wrote: post #152
bombcar wrote: post #114

Earlier quoted context omitted.

I suspect many of the “bad” responses are smart aleks saying “11111111” is just as likely a “62536164”.

111111 is just as likely as any other number. However, in practice, humans are far more likely to think of 111111 than other numbers, so we exploit the difference between the probability a human guesses a number vs. the probability of rolling a number on a fair dice. "Did I just roll a sequence X or am I lying?" vs. "What is the probability I roll sequence X next?" are quite different questions, if by "lying" you mea…

111111 is just as likely as any other *SEQUENCE* of numbers. This is a little confusing because the "sequential" requirement is somewhat masked by the repeating sequence used by the example. however, the odds of rolling six 1's is 0.00002143347.

considering the non-sequential set of rolls 625631 you have the odds of exactly two 6's at 0.201 and also one 2 5 3 and 1 each at 0.402

0.402^4 * 0.201 = 0.00524928641, or ~244 x more likely.

Re: We think this cool study we found is flawed. Help us reproduce it

#172

Post by simias »

marcosdumay wrote: post #140
simias wrote: post #135

Earlier quoted context omitted.

>It's always impossible to tell, for any given sequence, whether it was produced by a fair die. If the sequence is long enough you can model how likely it is to have been produced by a fair die. Are all numbers equally distributed? Are some numbers more likely to follow or not follow other numbers? Are some patterns repeating? Of course any sequence can be produced by a fair die, but you can still create some objecti…

Any single sequence of numbers has the exact same probability of being produced by a fair die (that's how the definition of "fair" goes). The probability of getting all 6 is the same of any other one you get.

For sure, but I think it's more helpful to this about "classes" of sequences. Sequences which have a uniform distribution of digits (within some margin of error), sequences which do not have repeating patterns, sequences that do not contain the same digit twice in a row etc... Any single one of these sequence is as likely as any other, but some "classes" are vastly bigger (and therefore, more probable) than others. By deciding which classes of sequence any result belongs to, you can decide if it's likely to have been produced by a fair die or not.

This intersects with the concept of entropy: assuming that you have a box containing a gas whose particles move randomly about the volume of the box, then at some point you take a snapshot of the position of every single particle in the box and you discover that they're all in the right half of the box, the left side being in a vacuum. Would you assume that it's just random chance? It could be. It certainly isn't.

Meanwhile any of the trillions and trillions of snapshots showing particles more or less uniformly distributed within the box are all more "random looking" and are what is expected from such an experiment. These configurations as a group occupy the vast majority of the phase space for the contents of the box.

Re: We think this cool study we found is flawed. Help us reproduce it

#173

Post by photochemsyn »

somehnacct3757 wrote: post #160

If you're forced to pick random numbers between 1 and X in your head, pick instead from a wider range of numbers and then modulo X. Your brain will legitimately have no idea what number you're picking. e.g. for a range 1-6, pick from 100-250 instead and modulo 6 plus 1. There are of course brand new biases at play (is your new range cleanly divisible by X?) But it's enough to tamp down the original biases you're worr…

I think you'd be better off taking a small pinch of sand, salt, pepper etc., throwing that on a smooth surface, then counting all the grains and then modulo it (just have the number of grains be >> than the range as in your example). This would reduce a lot of inherent biases, although perhaps introduce others.

Reminds me of the Buttered Toast Ig Noble Prize:

https://gizmodo.com/an-experiment-that-solves-the-worlds-mos...

Re: We think this cool study we found is flawed. Help us reproduce it

#174

Post by helloworld8 »

Not a scientist but perhaps one can help me understand, wouldn’t posting this to a forum with many people interested in software add potential for selection bias. I’m guessing that people like this are more inclined to understand what a “random” sequence might look like and therefore skew the results. If not, why not?

Re: We think this cool study we found is flawed. Help us reproduce it

#175

Post by zarriak »

elif wrote: post #153

I believe this is conflating distribution with randomness. Having played too many games and rolled too many real and pseudorandom dice, i know streaks are the rule not the exception, and that completely missed rolls likewise expectable. Using this model of randomness, i tried to create sequences that matched. The result is it says i have 60 year old brain. I did the prompt again, but simply /ensured there were no mis…

It would be nice if they had a different page where you could see in real time or after you submit a sample of 10 how random their model thought the inputs you were giving were.

Re: We think this cool study we found is flawed. Help us reproduce it

#176

Post by johnday »

photochemsyn wrote: post #173
somehnacct3757 wrote: post #160

If you're forced to pick random numbers between 1 and X in your head, pick instead from a wider range of numbers and then modulo X. Your brain will legitimately have no idea what number you're picking. e.g. for a range 1-6, pick from 100-250 instead and modulo 6 plus 1. There are of course brand new biases at play (is your new range cleanly divisible by X?) But it's enough to tamp down the original biases you're worr…

I think you'd be better off taking a small pinch of sand, salt, pepper etc., throwing that on a smooth surface, then counting all the grains and then modulo it (just have the number of grains be >> than the range as in your example). This would reduce a lot of inherent biases, although perhaps introduce others. Reminds me of the Buttered Toast Ig Noble Prize: https://gizmodo.com/an-experiment-that-solves-the-worlds-m…

I'm not sure that counts as "in your head", though. If such things were allowed, just look at your watch and mod the seconds by 6 (or whatever).

Re: We think this cool study we found is flawed. Help us reproduce it

#177

Post by alisonkisk »

denton-scratch wrote: post #156
HWR_14 wrote: post #150

Earlier quoted context omitted.

> It's always impossible to tell, for any given sequence, whether it was produced by a fair die. There's nothing an experimental subject can do to make the impossible more impossible. That's just not true. Or feel free to play a game with me. We'll roll a 20 sided die. If it comes up 20, you give me a ten. If it comes up any other number, I'll get you a dollar. Nice EV on that! Oh, the die has come up 20, 20, 20, 20,…

I would; would you? The Gambler's Paradox says you shouldn't (I'm assuming your 20-sided die isn't crooked). Incidentally, you haven't made your case that you can ever tell whether a given sequence was produced by a fair die. You've just asserted it, and then suggested a game that doesn't illuminate anything.

No post body was provided.

Re: We think this cool study we found is flawed. Help us reproduce it

#178

Post by johnday »

oneoff786 wrote: post #161
whatever1 wrote: post #159

Earlier quoted context omitted.

B-but statistics taught me that the events are independent! Take my 10!

The prompt told you it was independent. Not statistics.

Neither the prompt nor statistics told anyone it was independent. The prompt just says a 20-sided die, not a fair one.

Re: We think this cool study we found is flawed. Help us reproduce it

#179

Post by alisonkisk »

gus_massa wrote: post #151
simias wrote: post #135

Earlier quoted context omitted.

>It's always impossible to tell, for any given sequence, whether it was produced by a fair die. If the sequence is long enough you can model how likely it is to have been produced by a fair die. Are all numbers equally distributed? Are some numbers more likely to follow or not follow other numbers? Are some patterns repeating? Of course any sequence can be produced by a fair die, but you can still create some objecti…

> Can you roll a fair die a thousand times and only get 6s? Anecdote time: A few years ago with my friends we were discussing how easy is to roll 5 dices simultaneously and get the same result in all of them. This is a possible way to win a popular game here https://en.wikipedia.org/wiki/Generala We estimated how often you can roll the dices and the probability, and we estimated that you must try during 2 or 3 hours…

Saving a click: Generala is an ancestor of Yahtzee.

Re: We think this cool study we found is flawed. Help us reproduce it

#180

Post by helloworld8 »

photochemsyn wrote: post #173
somehnacct3757 wrote: post #160

If you're forced to pick random numbers between 1 and X in your head, pick instead from a wider range of numbers and then modulo X. Your brain will legitimately have no idea what number you're picking. e.g. for a range 1-6, pick from 100-250 instead and modulo 6 plus 1. There are of course brand new biases at play (is your new range cleanly divisible by X?) But it's enough to tamp down the original biases you're worr…

I think you'd be better off taking a small pinch of sand, salt, pepper etc., throwing that on a smooth surface, then counting all the grains and then modulo it (just have the number of grains be >> than the range as in your example). This would reduce a lot of inherent biases, although perhaps introduce others. Reminds me of the Buttered Toast Ig Noble Prize: https://gizmodo.com/an-experiment-that-solves-the-worlds-m…

At what point do we draw the line and say that these methods are sufficiently random? The task at hand is to come up with a sequence you perceive as random based on the numbers themselves, so adding layers like this seems to go against the concept of the experiment entirely.

How is this different from opening up my JavaScript console and doing Math.random() several times?

Post reply on HN