• Log in or sign up for an account here: https://www.sythe.org/login

Artificial Intelligence - The Future of Botting

Status
Not open for further replies.
I would use nothing but RiD if it wasn't for the auth system. No lifetime auth is fail.
 
I know what fuzzy logic is.
But in the end, its all boolean (if you like it or not).
And, fuzzy logic (the way you look at it) is possible in a script.

Yes, If you break it down to it's base level it is boolean, that's just the binary nature of computers. I think that we may be arguing over a miss understanding of terms.

From now on lets assume that Scripts using conventional programming methods do exactly what they are programmed to do and no more.

Artifical Intelligence is able to teach itself and emulate an intelligent response to the use.
 
It is very cool. I can't wait for the day when I can tell a computer to do something the same way I can talk to a person.

Still along ways away from having a desktop computer that can pull this off in a reasonable amount of time.

Just so you know the kind of power that took. It would take the fastest desktop computer 2-3 hours to answer one of those questions. The Watson super computer does this in 3-5 seconds.
 
Yes, If you break it down to it's base level it is boolean, that's just the binary nature of computers. I think that we may be arguing over a miss understanding of terms.

From now on lets assume that Scripts using conventional programming methods do exactly what they are programmed to do and no more.

Artifical Intelligence is able to teach itself and emulate an intelligent response to the use.
True.
Now, can you teach the RiD bots anything?
I cant, it looks all pre-programmed to me...
 
See that's where you get off. Machines can learn and essentially gain intelligence. Although it is taught (programmed) how to learn it can learn from it's mistakes and teach itself a new pastern. At the most rudiment it can use trial and error to find the best path.

I recommend you read up on machine learning, which is a fascinating area of computer science. Basically it allows computers to evolve behaviours based on ever changing data coming from its surrounds.

Here is what it can do that a normal script can not. A normal script can be told if it hits a road block turn around move 5 spaces to the left and attempt to execute script A again...loop until successful. The script will most likely eventually be successful but repeat the error again...essentially it has learned nothing.

A bot using A.I will of-course make mistakes, but it will log crucial data telling it the circumstances that caused it to be unsuccessful. It will continue to look for a way to complete the task learning from it's successes and failures as it goes along. In theory it should eventually reach a point of perfection although I don't personally believe this is possible (there are way to many variables to consider).

There are even some machine learning algorithms that were developed that allow computer programs to write other unique computer programs without they themselves being told (programmed) what to write. This is called algorithmic inductive inference and I believe was developed by E. Mark Gold.

I am not saying that there are not limits to machine learning the Turing Machine illustrates this more than clearly. What I am saying is that it is possible to at least mimic intelligence on a mechanical level without scripting every possible combination as you have mentioned
 
See that's where you get off. Machines can learn and essentially gain intelligence. Although it is taught (programmed) how to learn it can learn from it's mistakes and teach itself a new pastern. At the most rudiment it can use trial and error to find the best path.

I recommend you read up on machine learning, which is a fascinating area of computer science. Basically it allows computers to evolve behaviours based on ever changing data coming from its surrounds.

Here is what it can do that a normal script can not. A normal script can be told if it hits a road block turn around move 5 spaces to the left and attempt to execute script A again...loop until successful. The script will most likely eventually be successful but repeat the error again...essentially it has learned nothing.

A bot using A.I will of-course make mistakes, but it will log crucial data telling it the circumstances that caused it to be unsuccessful. It will continue to look for a way to complete the task learning from it's successes and failures as it goes along. In theory it should eventually reach a point of perfection although I don't personally believe this is possible (there are way to many variables to consider).

There are even some machine learning algorithms that were developed that allow computer programs to write other unique computer programs without they themselves being told (programmed) what to write. This is called algorithmic inductive inference and I believe was developed by E. Mark Gold.

I am not saying that there are not limits to machine learning the Turing Machine illustrates this more than clearly. What I am saying is that it is possible to at least mimic intelligence on a mechanical level without scripting every possible combination as you have mentioned
All true, but I honestly don't believe that RiD made a AI like this and used it in a RuneScape bot (Which is also cloud based). In fact, if you look at the programmers @ IBM in your vid, they would pay RiD so much for his work that he'll never have to work again.

But hey, theres no AI like that in RiD.
Every time I say theres no AI in RiD you say it uses fuzzy logic, which is also used in scripts.
 
I do believe RiD uses fuzzy logic as method of learning about it's environment and I think they use machine learning algorithms to make the bot more efficient. They don't have to be super powerful and honestly you see the technology everywhere...For example most modern search engines implement machine learning based algorithms to learn what you want before you type it based on previous queries. Sites like Mapquest implement these methods to determine the fastest route of travel...If it can be implemented there why is it so hard to believe that it can't be implemented in an RSbot?

Also you mentioned it's cloud based hinting that this would further limit the power of the bot to execute these algorithms... Obviously RiD has more processing power server-side making this the most efficient method. If they were executed locally it may severely slow down ones computer.
 
I do believe RiD uses fuzzy logic as method of learning about it's environment and I think they use machine learning algorithms to make the bot more efficient. They don't have to be super powerful and honestly you see the technology everywhere...For example most modern search engines implement machine learning based algorithms to learn what you want before you type it based on previous queries. Sites like Mapquest implement these methods to determine the fastest route of travel...If it can be implemented there why is it so hard to believe that it can't be implemented in an RSbot?

Also you mentioned it's cloud based hinting that this would further limit the power of the bot to execute these algorithms... Obviously RiD has more processing power server-side making this the most efficient method. If they were executed locally it may severely slow down ones computer.
Chat bots and GPS systems have been out there for a long time, but a GPS aint an AI and a chatbot is far the same as you claim RiD to be (Its the same as the auto chat responder, which is very silent over here).

There's a GPS implented into Nexus so yea, thats all possible. But letting the bot think about how it can fix its broken pickaxe is not done.

About the cloud computing; I didn't say it would limit the abilities of the bot, I meant that it would be very impressive.
 
Google most certainly has implemented this: http://bit.ly/9ZmV7Y

And MapQuest and Google maps have used learning algorithms ever since clearview came out in 2008: http://bit.ly/gXxeIE

And your misunderstanding my analogy of GPS using A.I learning algorithms. Nexus uses a script that moves you from coordinates A,B to coordinates C,D using a web. Modern Direction finders use learning algorithms that determine the best path based on emperical data, which is always changing.

It's not a perfect comparison but it illustrates the basic concept of Machine Learning Algorithms.
 
Google most certainly has implemented this: http://bit.ly/9ZmV7Y

And MapQuest and Google maps have used learning algorithms ever since clearview came out in 2008: http://bit.ly/gXxeIE

And your misunderstanding my analogy of GPS using A.I learning algorithms. Nexus uses a script that moves you from coordinates A,B to coordinates C,D using a web. Modern Direction finders use learning algorithms that determine the best path based on emperical data, which is always changing.

It's not a perfect comparison but it illustrates the basic concept of Machine Learning Algorithms.

Modern Direction finders also work with coordinates and calculate the best possible path. This is what happens in nexus, it calculates the best bath, using agility shortcuts, the grand tree teleport, fairy rings, and so on.

And, I don't see why you need emperical data in runescape?
It's not that you can stand in front of a red light or anything...

The thing is, in runescape bots, scripts can get as AI as RiD's bots.


And the projects you keep mentioning are way out of RiD's lead.
 
Sorry for the long response. I was on vacation. I think me and you are going in circles here. You believe that anything a computer program does is determined by the programmer. I see where you're coming from, but I still disagree. With evolution and learning algorithms programs are no longer always 100% predictable. I'm of the school that believes computers will eventually be able to operate very differently than when they were originally programed. Like I mentioned above there are programs out there capable of rewriting themselves...which opens up a whole new can of worms...

Yeah, the programs I mentioned are a bit out RiD's league, but remember it's people like him that are often hired to develop those types of programs created by these companies.
 
Yeah, the programs I mentioned are a bit out RiD's league, but remember it's people like him that are often hired to develop those types of programs created by these companies.

And this is based on what study/story...?

In the video I see 40yo programmers that have been in the industry for years (hence IBM).
And no, a program written on a von neuman pc will never be unpredictable (look at olly).
 
I enjoyed reading this, nicely formulated and well put together. I knew most of what you stated, on a general basis, but you elaborated it nicely.

Thanks!
 
No, I was referring to the client. Not the entire game. I don't think you even understand the difference between client and server side so their is no point to explain it. I'd also like to point out that although the RS.jar (this time speaking about the whole game) is huge people have stolen the entire thing and hosted it on their own private servers. However reflection and bcel clients both download the entire RS *client* .jar which is about 30 mb.

Haha I read that comment and I said "I hope Jamin schools that kid...."

Good job schooling that kid.
 
Actually you're a fucking retard. Agility is by far one of the most useful non-combat skills in the game. Your run energy drains so slow, you'll never need energy pots or tireless run scrolls.

Faster and more efficient banking is the result of a great agility level.
And also RiD has a mining, second most demanded bot of all bots on the internet and RiD offers it for free! And it averages 35k xp+ an hour powermining iron so next time do some research before you post so you don't sound like a retard, kk?

What an awfully immature way to solve a disagreement. Fix yourself.
 
@Shane+



You can actually fix this problem by loading the bot up inside a virtual machine. They provide a free copy of windows xp preloaded with everything you need. If I were you I would designate 2 gig for the virtual image and 1 gig of video ram!

@thr0wback



No one is trying to prove you wrong and if you fully read my article you would see that I clearly differentiate between true AI and our attempts at AI. I also went into some detail about how neural nets work and computer learning. Also how do your uncles credentials make you anything other than a 3rd party source?

@Lamba


No, they don't need to dig into your computer your browser information, OS, ip address, screen resolution and all types of interesting stuff is broadcasts for just about anyone to see. If you don't believe me go here and see for yourself.

http://www.whatsmyip.org/more/

For me they were able to see the following information: User Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.2.13) Gecko/20101206 Ubuntu/10.10 (maverick) Firefox/3.6.13

Wait is that a lot of virtual image and video ram? Please cite your source :)

Woops my bad! I didn't see Jamin posted this, I thought it was a rando forum junkie... No, Jamin you don't have to cite your source lol you seem very knowledgeable about what you are talking about.
 
Status
Not open for further replies.
Runescape Gamblingrunescape gambling
Back
Top