Advertisement

Why is so hard to find anything useful?

Started by June 10, 2005 01:44 PM
66 comments, last by Coz 19 years, 4 months ago
Quote: Original post by DarkThrone
x = 7 is so great. I use it and now my machine can think, dream and solve the problem for dominate the world. Thanks.

Without sarcasm, I'm not questioning the DYNAMIC of a thing. A car is based in physic, with VELOCITY, TORQUE, RPM values. But a engine is built by MECHANICS, not PHYSIC. Even Mechanics uses Physics, it's relative to Mechanics the Physic formula used.

The problem is the theory, in some parts, show you anything so complex that you need a full mathemathical course rather than code. So, you finally know that this damned formula is more simple than 1+1. If in same theory, a simple step-by-step by example will teach you by association what that means.


Well, I'm an academic, so to a certain extent, I guess I value theory higher than code. My advisor once warned me that as long as I played only with code, I'd be just like any hacker that comes a dime a dozen. And I guess to a certain extent, one of the main reason I don't like giving out code is just that I don't want people falling into the same trap I stepped into. You can only build so much on what other people have made. A person can give you a recipe for a sponge cake, and all you'll ever be able to make would be variations of that. You can take apart the engine of a car, an gasoline engine, let's say V6, and modify it anyway you want, but chances are, you won't be able to build a Diesel engine.

Also, there's the effort factor. Somebody may have poured over the theory for hours or days to come up with the code, and they just don't feel that someone else should get it for free without going through the same effort. Its almost a pride thing. Personally, as a programmer, I believe that you never really learn anything until you build it yourself.

Quote:
Seems that I and Weirdofu agree with dream dynamic, but that's not the point. Since the base is psychology, first we have to gave a full structure of storage and manipulation sufficient efficient to dynamic runs out of errors. If you don't, you will seek and modify the dynamic code ( the mathemathical part ) to try repair a error that don't be there. Even you can sucess, it's not perfect, and in a point or another it fails. If you build a clear base, dynamic adjustements just enhance the project instead. That's the concept. The machine knows nothing, just you can teach it. If you are a good teacher, if will be a good program.


True for most solutions in the world and true for AI almost always, the Occam's Razor. In AI, sometimes and many times, things are ALOT simpler than you think. The question is rather, how do I break things down and simplify things to the extreme. Then, there is the issue of identifying, what is hard-wired and what is emergent. Alot of human behavior aren't hardwired or hardcoded, but rather by products of the original code. The original programmer may have no idea such results may occur, but it just does. And it is true that the machine knows nothing, but you shouldn't draw strict lines for it either. Even good teachers can't create good programs if the students are bad. :p
The Occam's Razor is a fake to a single that figures "The Lord of True". Into all possibilities, all are true as possibilities. Try each one is the game. Fail, fail and fail until find the right. Into all possibilities in programming a artificial intelligence, The Occam's Razor favorite is lemmon... just kidding.
Take serious, it's the "Ah, I never can make it. Is best sleep in my bed.". If you use the Occam's Razor at way-of-life, you are not a programmer.

The reason that I program that why I love it. Don't matter if my journey is faded to mess, it's what I want. Will Power is the first thing to survive, the key of life. Occam's Razor is the easy way to solve a problem.

Piece-by-piece I was finding that I need to build a dream. If this dream are what I search, I don't know. But it's the point. I take the same key, theory and code are both necessary to learn. If I want copy-and-paste, I will ask "Please, someone can help me because I'm a noob and want some copy-and-paste codes to make anything with sure will be a shit when I end."
Advertisement
Quote: Original post by DarkThrone
If IA is mathemathical, give me the magical formula of a dream. Or to be more simple, give me the formula that calculates the sensation when you awakes and acidentally smash the finger of your foot in the edge of the door.


Or perhaps, you could explain why dreams and sensations have anything to do with intelligence. They certainly have something to do with pschology, but only insofar as they help to define the 'human condition', which we have come to realise in the last century should probably be called the 'mammalian condition'. I doubt very much that insects dream (why would they need to... but it'd be an interesting universe in which they did!), but I certainly know of examples of intelligent insects. Or perhaps I should say, "insects that solve problems 'intelligently'"!

I find it disheartening when I read what some people write about learning new skills; that if someone would just show them the end result, they would be able to work it out for themselves. Learning is a process, not a result. Learning is about discovering 'how' to achieve something, not simply defining 'what' to achieve. If someone provides the answer for you, then you didn't learn anything... and I can guarantee that you'll expend more energy trying to unravel it by yourself than asking someone who already understands to explain it.

Having said that, I completely agree with ApochPiQ's comments; people become jaded with the online community. I'm a good example of that. When I first starting helping on this site I used to provide answers to many questions on a daily basis. I would write long, detailed explanations of techniques, provide algorithms for methods and sometimes even pseudo-code, so that people had most of the answer to their question. Understanding was provided, it just required implementation. After a few years though, I noticed that most people, especially beginners, didn't really want to know the latest, greatest (or best) AI technique for solving their problem. What they really wanted was for it to be already solved. Those that wanted to know how to solve something would just ask to be pointed to an appropriate resource. Today, I spend far less time in these forums. Partly because I now have many more committments for my time (work, family, friends, etc), but mostly because I've been around this block many times before. I'm certainly not going to do other peoples work for them, nor am I going to give them code that represents days, months and sometimes years of work by other people. I'm all for sharing knowledge about how to achieve something, but I don't support giving away peoples work (nor do I support breach of copyright). Hence, I don't provide code. If you're a competent programmer, you'll have no problems implementing an algorithm. If you're not a competent programmer, you shouldn't be training yourself on AI problems. That might sound harsh, but it's a fact of life.

Remember one thing as you go through life: understanding, not a collection of facts, is the true path to wisdom and knowledge.

Cheers,

Timkin
ufff... is so long topic, isn't it. Big, full of post, the great thing. I really love a post with much replies, show me that someone really matter.

If is general opinion about all, I really give up to try. Maybe all are right, but it's so frustrating, because I really figure that things don't was so.

I was recently seeing the forums and see very topics about the same thing, pathfinders, ANNs, genetic algorithm. Not really a single topic talking about fundamentals of human intelligence, oppinions, view points.

Maybe this topic really changes of your first point. I ask to a difficulty in search for results, and give me my view point of it. All reply and gave the view point of them. The discussion changes so for PHYLOSOPHY X MATHEMATICAL on IA implementation. It's so interesting. A real talk, it changes based in a query and most probably changes again when a new question is up.

I was thinking. The code really is not the secret to me. Some parts, for other side, is a big hole when I try to figure how I say to machine what I want to say. It maybe happen to you too, before 10~12 hours in front of a computer, thinking and doing all work.

Other day, I was looking for the life to try escape of this thing. See people with your problems, people sad, people happy, etc. So I understand.

The first step, the man can conquer a thing. He want, he do. Your objective it's your life. All impulses are turned to it, and max of your possibilities are to take it. He changes, abdicates, do all because it is the most important thing of the world and he does everything to do. This is the resume of my view point that was the WILL POWER.

All of us have the same WILL POWER. Some uses all targetting a single objective. Others divide it along very small objectives. But ever it's the same. If you figure that isn't true, look anyone that seems not will-determined and see that he dedicatates to small things a little. Look for a depressive and see that he's fight with himself because a wish ( don't matter which ) is so important to this that the single idea of fail have him sick.

It's not a critical for everyone, please. It's a view point that I use to base my IA implementation. Maybe the view point of mine is useful for any that will think in a will power implemetation ( maybe a motivation engine ).

If any want to share view points, feel at home.

Ah, a man is what a man knows. You knows, understand and wisdomly uses that.

And dreams are so important. Human beings are great creatures, with mental capabilities above all other living beings.
But this capabilities are based in a sense that a power flux that will be hard to control. Dreams are some a chamber in sections of tube, where this flux roll some times to less velocity and don't destroy the tubes on a curve. It happens to me. Magically, a answer for a problem just "light" on your head some months after, even that you can't think on it anymore. The brain uses all info and associate it. All living beings do so. Dreams are to complete the table, on spaces that things that never exists can be made.
Quote: Original post by DarkThrone
The Occam's Razor is a fake to a single that figures "The Lord of True". Into all possibilities, all are true as possibilities. Try each one is the game. Fail, fail and fail until find the right. Into all possibilities in programming a artificial intelligence, The Occam's Razor favorite is lemmon... just kidding.
Take serious, it's the "Ah, I never can make it. Is best sleep in my bed.". If you use the Occam's Razor at way-of-life, you are not a programmer.


Well, for starters, I don't think you understand the concept behind Occam's Razor. As a programmer, you need to use it all the time.

Occam's Razor - "one should not increase, beyond what is necessary, the number of entities required to explain anything", "one should not make more assumptions than the minimum needed"

You can read more here if you want. http://pespmc1.vub.ac.be/OCCAMRAZ.html

Thank you Timkin for the enlightening post. And as we are really diverging from the original topic, I do feel that what can be said and should be said about the original post has be stated.
Well, it's mine...

Sometimes I pass so much time in front my machine that forget some points and so I post things that sound as a suplic. Sorry about that.

I am not Lord of True. But I think that Artificial Intelligence is a "simulation" of real intelligence. And for evolutive chain, logic was grown over a feeling base. Maybe build a Logic Simulation and after this put feeling simulation can work, but maybe not.

Thanks for Thinkin to mark your presence here. He ask me what Dreams are to real intelligence. I said, maybe I can be wrong, important. Dreams are in part, a way that brain have to calculate new possibilities without the senses and councioness. If you stop and see, mind are ever active. As I said, sometimes a problem make months until be magically resolved by a glimpse idea. This idea, on true, was be tested many times in brain and finnaly it find the answer.

The TURING TEST was real proof of that. If the principle of Turing's Test is the machine simulate perfectly the human conversation, maybe if feeling was part of simulation no people can really discover who is the machine, and finnaly it's over.

Maybe I'm not using the right words, because I am not a good English Speaker.

Almost forget: Occam's Razor is the query of the day.
Well, I am think that,s questioning a scientifical phylosophy is "down the hill", so I can't now.

And yes, I agree, the original topic is over, but by the way it's direct goes to a new discussion. So if any want post your opinions, I really thanks.
Advertisement
I don't think it's hard to find things. It's just that you're not looking in the right places. This game site is the last one I would look up how our brain functions. I would scan medical sites first to gain fundamentals then look at programming sites to figure out how to code it. I would start with biology 101 that explains the cells and how they function, etc. Then look at medical logs that explain what happened to the subject when left half of the brain was cut off, etc. Basically, try to pose questions then experiment with objects to see how your ideas panned out. I don't have the stomach for this but it's fascinating science if you can get beyond the gross parts. If you want to learn how to paint you go to an artist. If you want to learn about our brain and AI you need to go to a doctor, or someone who has studied the brains of many animals, understands their physiology and learn from them. Our brain is a complex beast and there are many parts that work together in a certain way. You severe one small link and the brain is no longer intelligent. Heck, a mild chemistry imbalance and it spins it out of control. Why? Can it happen to your AI as well? Why not? How does child development influence the brain function? So a call to psychiatrist is in order as well to gain better understanding of why we are the way we are.
this is the first thing I do.
And if you see, this site is also a programmers resource.

And also, even games like GTA don't have a powerful IA. If programmers work more on game IA instead of Graphics, it'll more compatible eith old machines and more fun.
Quote: And also, even games like GTA don't have a powerful IA. If programmers work more on game IA instead of Graphics, it'll more compatible eith old machines and more fun.


This is a tough argument. The GTA games are massive sellers; while improved AI may make the game better for some people, it seems rather well liked as it is

Instead of asking the dev team to spend more time improving the AI, what specifically would you ask for? What would you like it to do that it doesn't do already? Or what would you want improved?

[Edited by - BrianL on June 18, 2005 10:39:51 PM]
Rockstar should focus on the gameplay issues which killed GTA3 for me. The AI was fine in that game I thought.

This topic is closed to new replies.

Advertisement