Artificial Intelligence

Reply #73

martor854 — Sun, 26 Sep 2010 00:36:34 GMT

Hi jimmy4164,

Thanks for your reply. I haven't made any changes to my browser and/or security software. I do not know why I can't log in as martor54. I follow the procedure, I get Log in successful but my user name is not there. However, I can do that no problemo under my slightly changed user name. I'll have to stick with that.

Best regards,

martor854

Reply #72

jimmy4164 — Sat, 25 Sep 2010 14:20:47 GMT

martor[8]54,

This entry in LotteryPost Help may provide a solution to your login problem.

https://www.lotterypost.com/help/62704

Jimmy4164

Reply #71

martor854 — Sat, 25 Sep 2010 03:58:37 GMT

Hi,

I'm the former martor54. I don't know why I can't log in as martor54 on the site.

Could you please shed some light on the matter? I can see you're good, very good. Any nudge in the right direction will be appreciated. Thanks.

Reply #70

bob790 — Sun, 19 Sep 2010 04:49:33 GMT

Thanks for replying Jimmy

Bob

Reply #69

jimmy4164 — Sun, 19 Sep 2010 02:37:10 GMT

bob790,

Most of your questions above are outside my areas of expertise. I am familiar with Shannon's analyses as they apply to changes in stock prices. However, believing that in the present, the probability of a past event is 1.0, I fail to see how his work would have much value applied to the classes of chaos/randomness produced by lottery ping pong ball machines or random number generators producing sequences with extremely long periods.

John Conover relies heavily on Shannon's work.... [ More ]

Reply #68

bob790 — Fri, 17 Sep 2010 22:40:49 GMT

Hi,

I wonder how Lotto can be defined. Is it a chaotic or a random or both space state vector? Is it linear or nonlinear, stationary or not stationary, deterministic or not, periodic or not periodic? Would the definition change if I use the drawing in grawing order or sorting order?

It is probably chaotic as the lyapunov exponent and shannon entropy are positive. I don't know for others parameters.

Defining the caracteristics of each lotto is important in choosing the right tools for an... [ More ]

Reply #67

jimmy4164 — Sun, 29 Aug 2010 16:30:17 GMT

In the case of inputs from lottery results, I believe the noise to signal ratio is VERY high. TOO high!

For daytrading, my vehicle of choice is QQQQ for Long and PSQ for short. I use a simple approach, no warrants, futures, or options. I'll occasionaly pair PSQ and QQQQ as a short term hedge, but generally, I rely on close scrutiny of my [one] trade, my indicators, and the kill switch! Even Kelley's Criterion allows for quite liberal percentages of capital in this market.

I'm an adherent o... [ More ]

Reply #66

jwhou — Sun, 29 Aug 2010 02:40:55 GMT

Well, I wasn't thinking of increasing the number of hidden layers but the very infrequent additional hidden layer element being asserted which has the effect of allowing the neural net to consider another avenue of thought. Of course such an assertion has a destabilizing effect and an excessive number of exerted elements in the hidden layer would be excessive curve fitting. Ideally, you want the minimum number of elements in the hidden layer exerted. I always wondered about that since many three... [ More ]

Reply #65

jimmy4164 — Sat, 28 Aug 2010 19:53:48 GMT

This doesn't agree with my experiences trying to find predictive value in data with at best marginal non-random elements. In such cases, to elicit any useful information from the inputs required reducing the number of variables and/or hidden layers, rather than increasing them. Otherwise, the result was curve fitting extrordinaire!

There may be a rationale of backpropagation of known random inputs and that is to try and average out the effect of any random components from actual trainin... [ More ]

Reply #64

jwhou — Sat, 28 Aug 2010 18:53:28 GMT

I would say that you have to assume that what has been learned from previous iterations had been of some value so the danger is in having that value negated by noise in subsequent iterations. Of course, noise should cancel out but that isn't a given. The assumption that previous back propagation iterations have value that you would not want to lose leads to the concept that each iteration should adjust the values by a fraction of the previous iteration. Intuitively I would suggest that the incre... [ More ]

Reply #63

jimmy4164 — Sat, 28 Aug 2010 03:21:01 GMT

But I'm still wondering about this question, which we both forgot about last night!

...but really would like you to address the issue of the reasonableness of Backpropagation of essentially random inputs, and overtraining...

Reply #62

jwhou — Fri, 27 Aug 2010 22:49:53 GMT

That's the interesting thing about neural nets. We never really know what it finds out. At various times, people would try to work backwards and try to determine what logic the neural net has developed but of course, it could be anything.

With regards to non-linear programming. I was really hoping to avoid doing a 38 dimensional gradient for a min max Newton method solution. Just the very mention of 38 dimensions turns my head into a pretzel. It may all be moot anyways, although the method I... [ More ]

Reply #61

jimmy4164 — Fri, 27 Aug 2010 21:17:41 GMT

Another interesting indicator to factor in would be the product of the previous day's average minus close (I was going to say open minus close but as I've mentioned, the open has an artificial element hence perhaps an average would be more appropriate) and the volume to see if there's an overall consumer based momentum.

Interesting idea to avoid the Open ambiguity. When I'm optimizing 2 or more parameters I assume the network software is sorting out their products with at least one hidden la... [ More ]

Reply #60

jwhou — Fri, 27 Aug 2010 14:06:28 GMT

Sorry, I thought you were trying to predict the open. The day's open minus yesterday's close is an interesting index as it's an indicator what the latent demand is. The market maker would've selected the open such that it would sweep through any orders, stops and limits placed so as to maximize his brokerage's commissions. This would give you the initial day's momentum which may very well define what the day's high or low would be. Another interesting indicator to factor in would be the product... [ More ]

Reply #59

jimmy4164 — Fri, 27 Aug 2010 06:20:40 GMT

Made the mistake of peaking after logout!

My High/Low predictor model uses the (Open minus Yesterday's Close), so I KNOW the Open.

--Later

Reply #58

jimmy4164 — Fri, 27 Aug 2010 06:04:11 GMT

I've been sigmoiding all my inputs to [0...1], but your idea of segmenting that range may have merit for certain variables. We're getting into the Fuzzy Logic realm here. I also tend to train on 75% of the input and test on 25%, somewhat arbitrarily. Once a viable model is attained, I like to retrain as often as possible with up to date input.

Must turn in now but really would like you to address the issue of the reasonableness of Backpropagation of essentially random inputs, and overtraining... [ More ]

Reply #57

jwhou — Fri, 27 Aug 2010 05:54:18 GMT

The opens are set by designated market makers for each stock with access to information at the brokerage houses that you don't have. They are set to sweep the buy sell orders already placed in the system. It would be difficult for a neural net to predict the opens as it is essentially estimating an intermediate latent demand for transactions that can never be known. Then again, perhaps that's why a neural net would be used when all the inputs cannot be quantified.

Reply #56

jwhou — Fri, 27 Aug 2010 05:47:59 GMT

For investments, I suggest Shannon and Thorp. It's not neural nets but it's brilliant.

Reply #55

jwhou — Fri, 27 Aug 2010 05:45:46 GMT

I find that backpropagation often becomes tricky when the information representation has value rather than just the boolean of it exists or not. This is because it's essentially tweaking a constant that's multiplied to that value hence becoming a gain on a signal. The consequence is that you're winding up with an analog computer and not a neural net that deals with concepts. That was why I had the example of the three digits being represented by three elements, six elements or eight elements wit... [ More ]

Reply #54

jimmy4164 — Fri, 27 Aug 2010 05:13:00 GMT

I'm currently tinkering with several Backpropagation models attempting to predict the current day's high and low at the open ( after the gap) for several ETFs for daytrading purposes. The inputs, among other things, include fairly recent volume and price data for gold, foreign market action, and the issue of interest. I anticipated possible connections with cause effect relationships among these variables, and have found some. However, notwithstanding the possibility of ball machine wear or larc... [ More ]

Reply #53

jimmy4164 — Fri, 27 Aug 2010 04:31:30 GMT

In the PA Pick-5 that would have earned about $15. At $3 per day for 7 days, that's a $21 outlay.

That's better than the expected $10.50! (21 - 15) is a $6 loss, but I think of it as the cost of 21 chances on a nice Jackpot!

Keep testing, one week is not enough to draw any conclusions.

Reply #52

jwhou — Fri, 27 Aug 2010 04:30:44 GMT

when you talk about back propagation, you're talking about neural networks and connectionism. That's not the only kind of artificial intelligence, the other main branch is heuristics and is closer to traditional programming.

Connectionism is mathematically very simple, it's a matrix multiplication, back propagation is adjusting the elements of the multiplier vector up or down according to what the impact of that multiplier should've been. This results in a single layer neural net which is ac... [ More ]

Reply #51

RedToad — Fri, 27 Aug 2010 04:00:54 GMT

IN ONE WEEK I WON 2 TIMES WITH 2 NUMBERS AND ONE TIME WITH 3 NUMBERS.

Reply #50

jimmy4164 — Fri, 27 Aug 2010 03:38:09 GMT

How is the backtesting going?

https://www.lotterypost.com/thread/215821/1751417

Reply #49

jimmy4164 — Fri, 27 Aug 2010 03:35:52 GMT

HAWK,

Could you please explain further? I don't understand what you mean here. The link to my post that you were commenting on is here:

https://www.lotterypost.com/thread/215821/1754681

Thanks,

--Jimmy4164

Reply #48

pick4hawk — Thu, 26 Aug 2010 15:18:27 GMT

If your running things upon sequences the back draw period will vary. commercial software out there is not designed to work with several different types of sequences.

Quantity of the digits over periods will not get you vary far.

HAWK

Reply #47

jimmy4164 — Wed, 25 Aug 2010 19:32:10 GMT

I've tested this software, the only thing is that you must go through the painstaking work to try and test results with differents NN parameters like the learning rate,

As I suggested above, you could optimize parameters like the learning rate on one set of results, not including, say, the last 3 months. If the software works, it should work for at least a few Draws into the test period, the beginning of the 3 month period it has never seen before.

You said, I've tested this software,...... [ More ]

Reply #46

themagician — Wed, 25 Aug 2010 15:02:26 GMT

I've tested this software, the only thing is that you must go through the painstaking work to try and test results with differents NN parameters like the learning rate,

I think an interesting thing would be to analize parameters besides past drawing data

and that those parameters should be adapted to your local lottery

in fact as you say: the combination of statistics with AI

well, that's what I would like to investigate anyhow

Reply #45

jimmy4164 — Mon, 23 Aug 2010 05:17:54 GMT

Be careful using this software! I could design a neural network that would predict winners with astonishing acuracy, provided you only tested it on the Draws used to Train it. It would revert to a mediocre 50/50 picker the moment you turned it loose on Draws it never saw before.

Be sure to FORWARD TEST the parameters you develop on paper BEFORE you throw any real money at it.

Reply #44

RedToad — Mon, 23 Aug 2010 03:20:42 GMT

Here is the best neural net I found, you can set it up for cash five. Out of 3 sets you get two numbers in one set.

Many times you get one number in each set.

The problem is to find the correct setting for the AI.

Here it is, http://stonisa.info/hobby.php

Go to number one

# Lotto Artificial Intelligence

Excellent software that uses Artificial Neural Network to try to predict numbers to be drawn in the current round using history of previous draws. The prediction

Reply #43

ntdu — Fri, 20 Aug 2010 07:44:21 GMT

Can I discuss something to you?

I have done alot of ANN softs to solve problems in variety fields. In lottery I also try 2 tools: excel add-in tools (which can be downloaded from neuroxl dot com) and labview soft (i creat it), but the results, in any case, is not expected.

When I bring up the level of ANN network to higher level, the situation can be promoted, but not much. In my experience, I think ANN operates in current computers, can not solve the lottery problems.

Reply #42

themagician — Tue, 03 Aug 2010 13:09:43 GMT

Yes, while other people play Sudoku I play with my numbers, lol

Reply #41

Delta Draw — Thu, 29 Jul 2010 08:02:17 GMT

Good for you!

Looking at the results and what can be is fun. Numbers are fun. You really got to love the challenge and that is part of the game, the entertainment. Not a bad hobby at all.

DD

Reply #40

themagician — Wed, 28 Jul 2010 14:39:24 GMT

Hi,

Thanks to every one for taking the time to talk about all this. I would like to ask Jimmy if it is not AI, what do you use to play? Any system? There is always a danger to dellusion when comming to something like lotto game.

But I see it like playing with numbers, that is already fun to me, some times I've made small amount and there is always the possibility for the big hit. I use AI sometimes, it is not the only thing though.

As for me I will continue with AI investigation, experi... [ More ]

Reply #39

jwhou — Wed, 28 Jul 2010 02:22:16 GMT

If you are playing with Neural Networks, you don't actually need any fancy software to do so. You can model a neural net with matrice multiplications.

Reply #38

Delta Draw — Mon, 26 Jul 2010 19:18:58 GMT

Q1: yes

Q2: yes. Game dependant.

As for randomness, I am of the school that choas seeks order.

DD

Reply #37

MillionsWanted — Sun, 25 Jul 2010 12:05:57 GMT

I don't believe AI will be able to pick numbers with high accuracy . If we could be able to make it come up with three hits once in a while on a pick-6 lottery we should be very pleased.

AI used in combination with statistics and a system will perhaps increase the posibillity of getting four number hits or five number hits more often.

Reply #36

jimmy4164 — Sat, 24 Jul 2010 21:47:46 GMT

There seem to be 2 schools of thought here on the predictability of Lotto games. Since Double-DUH, er, Delta Draw didn't specify whether he was confused by my questions, thinks they were totally irrelevant or naive, or otherwise not worth commenting on, I'm not sure which school he fits into.

The first school consists of those who are firmly entrenched in their belief in the Gambler's Fallacy. (See my other posts or check it out at Wikipedia.) If the selection of winners in a given Lotto game... [ More ]

Reply #35

Delta Draw — Sat, 24 Jul 2010 07:45:54 GMT

Double DUH,....

DD

Reply #34

jimmy4164 — Sat, 24 Jul 2010 05:22:41 GMT

Don't most states routinely swap out machines and sets of balls after each draw?

Also, do you really think 10 or 20 draws is a sufficient number to gain statistical significance for at least the 95% level, so you could have confidence in your analysis

Reply #33

MillionsWanted — Fri, 23 Jul 2010 22:17:12 GMT

The pattern might consist of repetitions, how often a number has been drawn in a perhaps 10 or 20 draws, how often a number has been drawn together with another number(s) etc.

This might be because of wear and tear on the drawing equipment and balls and/or how the drawing equipment and balls was designed and manufactured.

Reply #32

jimmy4164 — Fri, 23 Jul 2010 17:41:26 GMT

A lottery drawing pattern got nothing to do with balls communicating or even the balls remembering which numbers was drawn in the last draw.

What sort of patterns are you referring to then, and WHAT do you believe is causing them

Reply #31

MillionsWanted — Thu, 22 Jul 2010 11:20:35 GMT

A lottery drawing pattern got nothing to do with balls communicating or even the balls remembering which numbers was drawn in the last draw.

For a period I tested the Excel addon PredictorXL on lottery draws. I was only using dates and the numbers drawn and very often it came up with one number correct, two numbers once in a while and three numbers so seldom it wasn't worth trying. I am sure there are methods of using NNs to improve on my results. There are several ways of implementing NNs. P... [ More ]

Reply #30

RL-RANDOMLOGIC — Thu, 22 Jul 2010 07:18:40 GMT

Jimmy

I understand what you are trying to do and maybe this is the best place to do it. I have a different

view of random theory when it comes to lotteries but agree that this might be a problem for some.

RL

Reply #29

jimmy4164 — Thu, 22 Jul 2010 03:44:20 GMT

RL,

Would you believe, I just inadvertantly included a Link in a reply to you that I spent over an hour on? I dealt with most of the issues in your post but I don't have the time or energy to try to re-compose it now. Briefly, I took back my accusation of propagating ignorance but re-asserted my Homework recommendations of analysing the PA Daily Number 33 year results and the Gambler's Fallacy at wikipedia. I said I think so-called Lottery Systems give false hope to poor people causing them t... [ More ]

Reply #28

RL-RANDOMLOGIC — Wed, 21 Jul 2010 23:11:20 GMT

jimmy

What I don't understand is how you can invest $6.00 for your entertainment and then turn around

and and accuse me of propagating ignorance . Did you never consider that I do this because I love

the chase. The quick picks that you buy come from a system. This system has been designed to pick

a random set of numbers. Consider this, How much bias is used to make your QP's random. You

cannot have any bias and not call it a system. What If I wrote a program that picked random set... [ More ]

Reply #27

jimmy4164 — Wed, 21 Jul 2010 21:22:36 GMT

You're disingenuous RL. If the Gambler's Fallacy was common knowledge here,

software that banks on people not understanding it would not be pushed here.

I play the Powerball using Quick Picks because I feel the $6 per week I spend provides

me with at least $6 worth of entertainment, with the added bonus of the possibility

of a big jackpot. What I don't do is waste time and money trying to beat the odds under

the mistaken belief that ping pong balls can talk to each other. I though... [ More ]

Reply #26

RL-RANDOMLOGIC — Wed, 21 Jul 2010 20:16:41 GMT

jimmy

I don't have to read it, It is common knowledge. I also do not have to view the PA data.

Let me ask you why you are here, do you have the urge to expound unknown wisdom

that the rest of us are lacking or are you just regurgitating something you just learnt

watching TLC. I could say But tell that to the person holding the winning ticket Life

is a gamble. Everyday you go out and get in your car and drive to wherever, you are

risking being killed of injured but you still do... [ More ]

Reply #25

jimmy4164 — Wed, 21 Jul 2010 19:23:18 GMT

Sorry,

I confused things by making an error in the instructions for the PA Lottery. The first choice you must make is to choose from the list of Games, the Daily Number . The above works after that. Sorry.

--Jimmy

Reply #24

jimmy4164 — Wed, 21 Jul 2010 18:54:38 GMT

RJOh,

You said, I thought the idea of using AI was to look at data with no perceived ideas at the beginning and allow AI to develop its own perceptions. This is absolutely correct. However, the kinds of parameters you allow an AI program to try to make associations among is KEY. What assumptions are you going to make about the effect, if any, the previous results of a particular game could possibly have on future results? If you are currently designing Neural Networks you must be aware of the... [ More ]

Reply #23

jimmy4164 — Wed, 21 Jul 2010 18:20:48 GMT

RL,

I suspect you didn't take my advice above and study the Gambler's Fallacy at wikipedia.

You also need to understand the concepts of Variance, Standard Deviation, and Confidence Intervals. All of your comments above refer to short term results and based solidly in your belief in the Gambler's Fallacy. If you REALLY believe a particular game system in your state is currently biased one way or another, you should contact your state Representatives and call for an investigation! Before you... [ More ]

Reply #22

RL-RANDOMLOGIC — Wed, 21 Jul 2010 11:30:43 GMT

jimmy

True, trying to predict what the next set of numbers will be based on what happened in the last drawing

would be useless. lets say I find a group of numbers that has twice as many winning tickets as any other

any other group containing the same amount of sets. If I played from that group based on the odds I have

just as good of chance of hitting as I would any other group. However if the group continues to out perform

the other groups then I will have a better chance of hitt... [ More ]

Reply #21

RJOh — Tue, 20 Jul 2010 22:49:43 GMT

I thought the idea of using AI was to look at data with no perceived ideas at the beginning and allow AI to develop its own perceptions. You make it sounds as if AI looks at data with perceived ideas and decide which ones are the most appropriate depending on trends and other information it detects in the data. In other words a repeatable occurance that has been seen before when similiar conditions have existed. Since exact combinations aren't likely to repeat the combination itself must be part... [ More ]

Reply #20

jimmy4164 — Tue, 20 Jul 2010 21:14:10 GMT

The point I was trying to make, which I think you are missing, is that there are leading indicators in the stock markets, things such as earlier results of foreign markets in different time zones, volume considerations, and the provable persistance of trends, all of which give Neural Networks a possible edge. However, when it comes to ping pong balls bouncing randomly in response to air currents, there is NO WAY to predict which one will emerge when the release button is pushed. Unless there is

Reply #19

themagician — Tue, 20 Jul 2010 12:42:21 GMT

Well, yes, let's say that the word accuracy is a big one, but at least you use NN to assist, does that mean to find new tendencies in market? I've heard that some finances specialist were able to predict the european crisis through their own NN software, at least the output marked the lower comming tendency, it was that day when I became interested in artificial inteligence.

Well, as for me it is like a game, a passion for unsolved mysteries, if I could just find those tendencies I will be... [ More ]

Reply #18

jimmy4164 — Fri, 16 Jul 2010 15:01:54 GMT

I have been using Neural Networks for years to assist in selecting stocks and thoroughbreds. If you or anyone else can devise a NN that predicts lotto results with high accuracy on the in sample data, I suggest you apply the parameters derived to an out of sample dataset. I am certain you will be disappointed. Ping Pong Balls have no communication skills.

Reply #17

themagician — Thu, 15 Jul 2010 13:30:25 GMT

Hi,

As for the software I use I am writing it myself, it's taking me some time though and if there are some good software at abbordable price I might be getting it. But see, for example, Neurolox, there must be some pre-processing of the data before using it, calculating the input parameters, and you cannot always do all that with excel formulas. (Well at least I don't know that much on excel macros)

For now I need to see first which are the good parameters or views to the problem, I am tr... [ More ]

Reply #16

Delta Draw — Thu, 08 Jul 2010 07:43:23 GMT

You must have a valid question on that one to get an answer. What software do you think is good in ANN?

DD

Reply #15

frenchie — Tue, 06 Jul 2010 19:38:03 GMT

Hi everyone,

Here are 3 excel formulas that I find, if that can help someone ? with diferent bits lenght, from number to binary ,but I don't have the other way arround, binary to number,if someone has it I'm interrested.

Here they are :

First one 32 bits :

=DEC2BIN((MOD(A1,4294967296)/16777216),8) DEC2BIN(MOD(A1,16777216)/65536,8) DEC2BIN(MOD(A1,65536)/256,8) DEC2BIN(MOD(A1,256),8)

Second one 18 bits :

=CONCATENATE(DEC2BIN(INT(A1/512),9),DEC2BIN(MOD(A1,512),9))

Thir... [ More ]

Reply #14

themagician — Tue, 06 Jul 2010 16:05:39 GMT

I was taking a look at Neurolox, very interesting, it has a predictor add-on, for example to predict the result of a dog race it take one dog past data: dog's weight, the time of the day, and for the past output it takes the dog's timing in past races.

As always for artificial intelligence all is a matter of parameters,

but how could I pick up the parameters for a Pick 5 data past drawings?

Let's take number 15 for example, maybe I could take the total number of times it came out, the n... [ More ]

Reply #13

themagician — Tue, 06 Jul 2010 15:32:09 GMT

Sorry, I was wrong 11010 is 26 and he is taking numbers in columns

Reply #12

Delta Draw — Tue, 06 Jul 2010 06:40:07 GMT

Neuroxl has an add-on for excel. Careful tho, they are in the old russia and have a lousy distributer last time I looked. Many sites for the home company (and names?) and one I think was In the Maldives.

The site that came up on a G search shows a big sale going on. Keep me informed and PM me if you find anything out about this stuff. It looked good and better than most I found or had, just never took the leap. Make a purchase with a visa gift card to limit any bad biz from the distributer, t... [ More ]

Reply #11

GASMETERGUY — Fri, 02 Jul 2010 20:38:00 GMT

Correct me if I am wrong but 1101 in binary would be 13.

Reply #10

GASMETERGUY — Thu, 01 Jul 2010 16:38:04 GMT

If you are using Windows, you have a calculator which will convert decimal into binary and vise-versa. Click on Start , then All Programs , then Accessories , and find Calculator .

On the left hand side you will see HEX , DEC , OCT , and BIN . Click on DEC , input a decimal number, then click on BIN . The transition is made.

To go from BINary to DECimal, click on BIN, input your binary number, then click on DEC. The transition is made.

Reply #9

themagician — Thu, 01 Jul 2010 12:14:50 GMT

Hi,

I think he is not taken in consideration the lowest value in the colum to calculate the sum, and so 26 is: 1 1 0 1 in binary, I think that that's how he does it.

It is another way to find patterns, you need to test on it, it is only a matter of continue to experiment on it,

Reply #8

mikemerkury — Wed, 30 Jun 2010 00:40:41 GMT

Can you explain how you calculate decimal sum on binary

Reply #7

GASMETERGUY — Tue, 29 Jun 2010 15:59:39 GMT

1 0 0 1 0 0 1 1 0 1 = 589

1 1 0 1 0 0 1 0 0 0 = 840

0 0 1 1 0 0 0 1 0 1 = 197

1 0 0 1 0 0 0 0 1 1 = 579

0 0 1 0 1 1 0 1 1 0 = 182

RL

Another method you might try is to see all those zero's and one's as a binary number. I took the liberty of calculating the decimal equivalent of each row of binary numbers.

You could also track the columns as binary number.

1 0 0 1 0 0 1 1 0 1 = 589

1 1 0 1 0 0 1 0 0 0 = 840

0 0 1 1 0 0 0 1 0 1 = 197

1 0 0 1 0 0 0 0 1... [ More ]

Reply #6

RL-RANDOMLOGIC — Sun, 27 Jun 2010 04:37:55 GMT

TM

Thats the idea. Hope you can make more of it than I did

RL

Reply #5

themagician — Fri, 25 Jun 2010 13:38:43 GMT

If I understand well you test if a column formed by a group of four digits is true or false for the next drawing. If your second group is true, then one or more of these numbers 5,6,7,8 should be comming...mmm..

One of the approaches I had was taking the entire game as a matrix, or a string, in your game it will be a 1 to 39, lets's say then that number 3 has come out then the third position of my array will be 1 true , and 0 false for all the other numbers that didn't came out. Well I've tes... [ More ]

Reply #4

RL-RANDOMLOGIC — Thu, 24 Jun 2010 07:15:18 GMT

TM

Here is what I done. I made 10 groups of 4 numbers for my 5-39. Group 10 only had 3 numbers

because it is a 5 of 39. Next each time any of the numbers from a group hit it would count as a

1 if the group missed it would be a 0 . each draw was converted into a binary string that looked

like this.

1 0 0 1 0 0 1 1 0 1 = 03-15-26-31-39

1 1 0 1 0 0 1 0 0 0

0 0 1 1 0 0 0 1 0 1

1 0 0 1 0 0 0 0 1 1

0 0 1 0 1 1 0 1 1 0

I done this for all draws in my database. Next I... [ More ]

Reply #3

themagician — Wed, 23 Jun 2010 13:21:51 GMT

Ok, I will take a look at Brain Maker, a little bit expensive though..

It is true you cannot input raw numbers to the network,and there are some aproaches here, either taking the whole thing as a 1-0 matrix, or to scale all values before input, ah, I keep on testing.. and I do everything on my sparetime, last time I've got 18-20-36.. the comming out numbers beeing 19-21-36.., and it is good, getting closer, I want to get some more

Reply #2

Newb — Thu, 17 Jun 2010 20:40:04 GMT

Found this link to AI programs and neural networks.

http://www.dmoz.org/Computers/Artificial_Intelligence/Neural_Networks/Software/

I have been 'tinkering' around with converting drawn numbers into binary patterns eg if the number 3 was drawn once over 5 draws the pattern may look like this F,T,F,F,F where T = drawn and F = not drawn. It is possible to find patterns that repeat and some are more frequent than others. But as always the problem knowing when the patttern will occur again... [ More ]

Reply #1

RL-RANDOMLOGIC — Wed, 16 Jun 2010 19:24:05 GMT

themagician

If you find something then please pass it on. I too have been working with AI and NN but as yet have

nothing to share. Have you ever looked into Brain Maker, if not then google it and read some of the

examples. I was most interested in the example of the binary srting prediction that was above 90%

after being trained. It is a bit high $$$ but I might give it a shot. backpropagation I think would be the

way to go concerning the lottery but I think before any real headwa... [ More ]

Artificial Intelligence

themagician — Mon, 14 Jun 2010 15:43:55 GMT

Hi,

I will like to know if somebody is seriously studying or using artificial intelligence for lotto, I know there are some old posts about it and some explanations but has somebody got good results using artificial intelligence software or algorithms?

Is important to me because I am currently working on this kind of software, using backpropagation and it will be very helpful to get some feedback from your experience.

Thank you