Would a Neural network improve dice betting?

[KickItDown]

Or it will just end up mimicking what an human would do to win?

[lite]

Or it will just end up mimicking what an human would do to win?

Improve dice betting how? i think neural network will probably mimic what we'll do. they will too lose in end like us. if only there was a model available to train and see how it goes, it would be fun to see.

[pixie85]

No computer can predict an outcome of a roll. It's totally random. A neural network or whatever won't change anything, when all a user can do is choose how much money they want to bet and press a button. In a game where there are more actions to perform like rising/checking/folding that would be a great improvement, but dice is too simple.

[NorrisK]

Dice is random, with a negative expected value.

All a neural network would achieve was playing perfectly, thus losing approximately the house edge on average.

[FrueGreads]

Or it will just end up mimicking what an human would do to win?

How can dice betting be improved? The AI would probably just see that whatever it did, the profit would still be negative. It would be nice to see if it would be able to get some good betting strategy on short term betting. Like what is the best approach for a month betting period. Like what is the type of betting that would minimize your losses on a short period of time.

[Wendigo]

I want a neural network that works for sportsbooks. Stone cold AI without a droplet of human emotion  Analyzing the games and betting on its own. Easy profit I want that one 

[Pettuh4]

I want a neural network that works for sportsbooks. Stone cold AI without a droplet of human emotion   Analyzing the games and betting on its own. Easy profit I want that one  

That's sounds like fun but I doubt if it could do better than statistical analysis of previous results of the teams involved. Since it's going to model the human brain and nervous system it is likely to analyse only the statistics without human emotions like you want it and that is what will be its limitation and so you cannot win always.

[KickItDown]

I'm getting obsessed by the idea of a neural network that just learns from the "mistakes".

[NorrisK]

I want a neural network that works for sportsbooks. Stone cold AI without a droplet of human emotion   Analyzing the games and betting on its own. Easy profit I want that one  

That's sounds like fun but I doubt if it could do better than statistical analysis of previous results of the teams involved. Since it's going to model the human brain and nervous system it is likely to analyse only the statistics without human emotions like you want it and that is what will be its limitation and so you cannot win always.

It would probably also take into account developments as they occur (e.g. injuries) much faster than any model available and can potentially jump on arbitrage options much better.

[beerlover]

Certainly we can experience lots of improvements due to adopting neural networks for gambling. But the catch is who is going to adopt first, you or house ? Only the early adopter may find the edge still I am sure in long run neural network also will not bring any big differences.

[Ewinsane]

Certainly we can experience lots of improvements due to adopting neural networks for gambling. But the catch is who is going to adopt first, you or house ? Only the early adopter may find the edge still I am sure in long run neural network also will not bring any big differences.

You are right. When the development is happening everywhere then there will be no significant advantage of using it neither in gambling nor in trading. Only when we are alone utilizing such new technology then there will be the possibility of finding big improvements in profit making kind of things with gambling.

But we cannot say gambling houses also will update themselves for neural network improvements. Then there will be no point of adopting it.

[darkangel11]

I'm getting obsessed by the idea of a neural network that just learns from the "mistakes".

Great, but how would it relate to online dice?
Let's put a supercomputer with a neural link to work and give him a dice site to beat, where all it can do is click "roll". Oh, I lost, roll again or not? Let's roll, Oh, I won, roll or not? All I can foresee is that computer getting addicted and stuck playing dice forever 

[bluefirecorp]

Or it will just end up mimicking what an human would do to win?

What you think about bots ? Are those bots copying just what a gambler do or anything beyond special ? First I guess that we need to consider about the capabilities of neural network and then we can predict about its performances with respect to gambling.

I am sure neural networks are known for artificial intelligence hence we can expect to some drastic changes still there will be no possibilities to break the secret seeds so that we can know out of a dice well in advance.

[Caladonian]

Or it will just end up mimicking what an human would do to win?

What you think about bots ? Are those bots copying just what a gambler do or anything beyond special ? First I guess that we need to consider about the capabilities of neural network and then we can predict about its performances with respect to gambling.

I am sure neural networks are known for artificial intelligence hence we can expect to some drastic changes still there will be no possibilities to break the secret seeds so that we can know out of a dice well in advance.

its been programmed for randomly so its real hard to break it and i'm sure house will make it secret as far as they can prevent possible hacked
many gamblers are looking for edge and possible bugs to take advantage.

[milewilda]

Or it will just end up mimicking what an human would do to win?

Its senseless and wont really work after all because even neural network cant really beat up house edge no matter what. It may not have an emotion but knowing dice its pure random and theres no way to manipulate it or exploit it which theres no possible way to have winning situation anytime.It might work for some other fields but not for dice for sure.

[boranes]

Neural network won't improve dice betting, as you said it will mimic what we do.
It can't improve it because result is random number.