|
France12463 Posts
On June 10 2017 08:24 Ernaine wrote: Well, I don't agree. Yes, dices may not be fair. No real phyiscal dice can be infinitely fair.
If we throw 2 times and get both results once, the hypothesis that the true nature of the dice is that we get 10% coins and 90% heads still fits the data. Yes, after 2 trials we do get exactly 50/50, and we can get similar results at 4 trials and all the other even trials, but we don't know with a high probability that the coins for sure are in fact 50/50 completely fair coins.
So yes, we need to go to infinity to exactly get 50/50. You can try it with a computer. (yes, it will have pseudo-random numbers, so it is still a bit iffy, just like having a completely fair dice ? That's not what I am saying. I'm saying that you can have the same number of evens as the number of odds before infinity, not that you can be sure of the probability before going to infinity, which is the definition of probability so it's true by definition :o.
|
On June 09 2017 07:03 JimmyJRaynor wrote: lots of moaning whining and complaining... prof speaks up.... and says...
"if i flip a coin 35 times and it only comes up heads 2 times i do not statistically adjust that result. i note it as a special and unusual case". what he was saying was "there could well be 33 morons in this room ... so shut up you idiots"
It's even more devastating, when you take into account, that a coin actually has 3 sides. Seems almost like that one lucky bastard, who got 100/100 tossed a coin and it came up on the 3rd side.
|
On June 10 2017 08:28 Poopi wrote:Show nested quote +On June 10 2017 08:24 Ernaine wrote: Well, I don't agree. Yes, dices may not be fair. No real phyiscal dice can be infinitely fair.
If we throw 2 times and get both results once, the hypothesis that the true nature of the dice is that we get 10% coins and 90% heads still fits the data. Yes, after 2 trials we do get exactly 50/50, and we can get similar results at 4 trials and all the other even trials, but we don't know with a high probability that the coins for sure are in fact 50/50 completely fair coins.
So yes, we need to go to infinity to exactly get 50/50. You can try it with a computer. (yes, it will have pseudo-random numbers, so it is still a bit iffy, just like having a completely fair dice ? That's not what I am saying. I'm saying that you can have the same number of evens as the number of odds before infinity, not that you can be sure of the probability before going to infinity, which is the definition of probability so it's true by definition :o.
Either what you are saying is wrong, or it is besides the point. Whatever the true (secret?) probability the coin is, there is always a possibility that after an even number of flips, you have exactly the same number of odds and evens. And the larger the number of throws gets, the smaller those odds are. Even in the case of a perfectly fair coin. Only towards infinity will the deviations from 50/50 slowly disappear.
If you refuse to take my word for it, try it out yourself.
|
France12463 Posts
On June 10 2017 08:49 Ernaine wrote:Show nested quote +On June 10 2017 08:28 Poopi wrote:On June 10 2017 08:24 Ernaine wrote: Well, I don't agree. Yes, dices may not be fair. No real phyiscal dice can be infinitely fair.
If we throw 2 times and get both results once, the hypothesis that the true nature of the dice is that we get 10% coins and 90% heads still fits the data. Yes, after 2 trials we do get exactly 50/50, and we can get similar results at 4 trials and all the other even trials, but we don't know with a high probability that the coins for sure are in fact 50/50 completely fair coins.
So yes, we need to go to infinity to exactly get 50/50. You can try it with a computer. (yes, it will have pseudo-random numbers, so it is still a bit iffy, just like having a completely fair dice ? That's not what I am saying. I'm saying that you can have the same number of evens as the number of odds before infinity, not that you can be sure of the probability before going to infinity, which is the definition of probability so it's true by definition :o. Either what you are saying is wrong, or it is besides the point. Whatever the true (secret?) probability the coin is, there is always a possibility that after an even number of flips, you have exactly the sane number of odds and evens. And the larger the number of throws gets, the smaller those odds are. Even in the case of a perfectly fair coin. Only towards infinity will the deviations from 50/50 slowly disappear. If you refuse to take my word for it, try it out yourself. ? I'm just saying that if you throw it 2 times you can have an even and a odd, how is that wrong? There is no point because you didn't answer my question and went on about the definition of probability, which I already know :x. edit: oh nevermind you are the troll from the other time, that's why you talk non sense all the time xD.
|
I talk nonsense because I have a PhD. That's what we do. Troll. Sorry bro. These things are no longer a matter of opinion.
The 'troll from the other time'? At least my account doesn't have a troll name.
[edit]
Ooh, I see now what you mean. You refer to the Go-Deepmind thread, where you casually called people idiots for not agreeing with your trivially wrong position. I called you out, and you didn't like that. How do people like you get 7000 posts on TL? Amazing.
If you throw a coin twice, and it lands 'heads' once and 'tails' once, you know absolutely nothing.
|
There's going to be a low threshold for disrespect or trolling in this thread. So don't do it. I am looking at the guy above me.
If there is a communication problem, then work to solve it. If you're frustrated with a conversation and can't reply with respect then don't reply.
This is your only warning. I am adding asking tofu to add a zero tolerance rule about being rude or off topic. I don't care about previous squabbles.
|
I am one of the few people here able to make quality posts. Why personally attack me for no reason? We have a guy here saying it is fair to believe a coin is 50/50 after throwing it twice. And he calls me a troll. What do you want me to do? Yes, I know not everyone has a sophisticated mind. But that is ok, as long as you can accept that. I am looking at the guy above me.
This my only warning.
|
Well I am talking to him too, then, since he is also breaking that rule.
I don't think this is a lot to ask from the type of people that would be participating in this thread.
|
Honestly, I think there's a lot of conversational equivocation going on right now between the law of large numbers in regards to theoretical probability and then the empirical, actual results obtained along the way. People are making very valid points about probabilistic models but are also talking past each other and getting hung up on semantics.
|
On June 10 2017 09:03 Ernaine wrote: If you throw a coin twice, and it lands 'heads' once and 'tails' once, you know absolutely nothing.
Well, absolutely nothing is a stretch ... you know that the probability of getting heads at the time you made the first throw and in the conditions of that event were non zero.
If you make the assumption that the outcome of a coin toss is multinomial distribution fixed over time, you can say that the list of outcomes contains the 'heads' and 'tails' events with a non null probability.
|
In reality the roll of dice or throw of a coin is perfectly deterministic. Sure we cannot calculate this (most of the time) given that there are multiple forces at play and their values/vetors are extremely hard to measure accuratly. Still such rolls are not really random in atrong sense.
|
That's the thing. Statistics, for a lot of things, are used for things that are deterministic in nature, but have too many parameters tied to them that it's almost impossible to obtain that deterministic outcome. Of course very educated guesses can be made using it, but it's rather sad that we -still- just can't perfectly model some deterministic events imo, or it's more of the fact that we have statistics and we're too comfy with going for an "one of the other" situation.
|
On June 10 2017 23:39 Silvanel wrote: In reality the roll of dice or throw of a coin is perfectly deterministic. Is it? Quatum mechanics say otherwise. You cannot now the position of every particle exactly or you change the outcome of the experiment.
|
Since when is a dice a particle?
|
On June 11 2017 04:32 Biolunar wrote:Show nested quote +On June 10 2017 23:39 Silvanel wrote: In reality the roll of dice or throw of a coin is perfectly deterministic. Is it? Quatum mechanics say otherwise. You cannot now the position of every particle exactly or you change the outcome of the experiment. That doesn't mean it's not deterministic. It just means we can't know all of the initial state.
|
On June 11 2017 04:48 CecilSunkure wrote:Show nested quote +On June 11 2017 04:32 Biolunar wrote:On June 10 2017 23:39 Silvanel wrote: In reality the roll of dice or throw of a coin is perfectly deterministic. Is it? Quatum mechanics say otherwise. You cannot now the position of every particle exactly or you change the outcome of the experiment. That doesn't mean it's not deterministic. It just means we can't know all of the initial state. Hence it is not deterministic.
|
Human inability != stochastic process
|
On June 11 2017 08:04 Biolunar wrote:Show nested quote +On June 11 2017 04:48 CecilSunkure wrote:On June 11 2017 04:32 Biolunar wrote:On June 10 2017 23:39 Silvanel wrote: In reality the roll of dice or throw of a coin is perfectly deterministic. Is it? Quatum mechanics say otherwise. You cannot now the position of every particle exactly or you change the outcome of the experiment. That doesn't mean it's not deterministic. It just means we can't know all of the initial state. Hence it is not deterministic. Einstein's special relativity disagrees with you, spacetime and every aspect of our universe are entirely predetermined.
Edit: You also might wanna reevaluate your point about quantum mechanics, "every quantum state can be represented as a sum of two or more other distinct states" and "the result will be another valid quantum state". (Source: Quantum Superposition)
|
United States10327 Posts
was excited to see a math thread but it's already devolved into philosophy
|
So I've spent a day studying Fourier Transform. So far the math is very understandable, at least on how to apply it. Understanding why it works and specifically what it does is more challenging. Most resources on the subject are just absolutely awful, and I may have to go out and buy and old text book to get good information.
It seems like FFT implementations popped up around the 80s-90s, and nobody ever really looked at them again in depth. Since then everyone takes and old implementation and wraps it. Well there's FFTW, but that library is a software engineering nightmare. It's a horribly obfuscated and heavyweight solution for a particularly small function. But what can you expect from a bunch of academics.
The real challenge is to be able to derive all the math and understand it in all its correctness. Only this way can I write a good implementation.
So, any book recommendations from anyone that really understands the math?
|
|
|
|