Programming w/ Artificial Intelligence

[Aleksandar]

Does your mother know what you are thinking about?
I am not allowed to talk about stuff like this, but,
I am guessing that it has something to do with drugs?
Yes , No , Close
15. Is it more valuable than silver? Yes.
14. Is it soft? Yes.
13. Is it annoying? No.
12. Can it float? No.
11. Is it green or black? Yes.
10. Does it get really hot? Yes.
9. Do you chew it? No.
8. Is it crunchy? No.
7. Does it come from a plant? Yes.
6. Does it have tough skin? No.
5. Is it brown? No.
4. Does it go in your mouth? Yes.
3. Can you find it in a church? No.
2. Does it move? No.
1. Does it bring joy to people? Yes.
It is classified as Unknown.

Marijuana wow man not bad at all

 

[pakigang]

Originally posted by: harrkev

Originally posted by: pakigang
Just a question, are there any tutorials on programming AI. It just can't be if then else thing. Just curious.

Please define "AI." If you are talking about symbolic processing, this is difficult.

On the other hand, there are simple schemes like genetic algorithms (PBIL is a good one) and optimization schemes (like Q-learning) that are good for simple tasks (including robots). These are good for having a robot learn how to avoid walls and such.

I can give you some links if you are interested.

**EDIT**
Things like this can be done in C, Pascal, Matlab, BASIC, etc.

Yup, I would really be interested in programming AI. Can it be done in C++ or Visual Basic or Visual C++ (Winapi32 or MFC)

 

[nortexoid]

Nice, 25 guesses for 'window' (category: other)

 

[Jeffyboy]

Ok, well, it got Sex.
hehe... purdy funny

Does your mother know what you are thinking about?
I am not allowed to talk about stuff like this, but,
I am guessing that it has something to do with sex?
Yes , No , Close
29. I guess that it is education? No.
28. Is it something you can purchase? Yes.
27. I guess that it is work? No.
26. Can you get information by using it? Yes.
25. Can you pet it? No.
24. Can you buy it? Yes.
23. Can it help you find your way? No.
22. I guess that it is power? No.
21. Is it worth a lot of money? Yes.
20. I guess that it is mathematics? No.
19. Do you know any songs about it? Yes.
18. Can it be heard? No.
17. Does it require specific knowledge to use it? Yes.
16. Does it contain words? Irrelevant.
15. Is it used to calculate? Irrelevant.
14. Can you play games with it? Yes.
13. Is it an emotion? No.
12. Is it used by a baby? No.
11. Does it store information? No.
10. Does it go in your mouth? Irrelevant.
9. Does a basketball player use it? Yes.
8. Is it colorless? Irrelevant.
7. Does it bring joy to people? Yes.
6. Can it be washed? No.
5. Could it be found in a classroom? Yes.
4. Is it manufactured? No.
3. Does it come in a box? No.
2. Is it colorful? Irrelevant.
1. It is classified as Other.

 

[harrkev]

Originally posted by: pakigang
Yup, I would really be interested in programming AI. Can it be done in C++ or Visual Basic or Visual C++ (Winapi32 or MFC)

Any language that supports math and arrays can be used.

Here is some interesting reading to get you started (found by Google):
http://www.stat.lsa.umich.edu/...nforcementLearning.ppt

http://computing.dcu.ie/~humphrys/PhD/ch2.html This page is rather math intensive, but is all theory. Implementing Q-learning can be done with just add and multiply.

Just Google for Q-Learning or Reinforcement Learning and do some reading.

If you want to spend some money, these are both excellent books:
http://www.amazon.com/exec/obi...81/104-8684146-7959963
http://www.amazon.com/exec/obi...77/104-8684146-7959963

 

[S0Y73NTGR33N]

Here's what it gave me for CPU/Computer chip. Not very introspective took it 21 questions.




20Q.net
I am guessing that it is a computer chip?
Yes , No , Close
21. Does it come in a pack? No.
20. I guess that it is a transistor? Close.
19. Can you play games with it? Yes.
18. Is it used for communications? Yes.
17. Is it straight? Irrelevant.
16. Is it black? Irrelevant.
15. Can you hold it? Yes.
14. Does it require specific knowledge to use it? Yes.
13. Is it dangerous? No.
12. Is it something you bring along? No.
11. Is it a metal? Yes.
10. Would you find it on a farm? No.
9. Is it multicolored? Irrelevant.
8. Does it have a hard outer shell? Irrelevant.
7. Do you hold it when you use it? No.
6. Can it scratch? No.
5. Can you play with it? No.
4. Is it smaller than a golf ball? Yes.
3. Does it help accomplish tasks? Yes.
2. Is it manufactured? Yes.
1. It is classified as Mineral.
?

 

[complacent]

Originally posted by: Jeffyboy
usually they teach AI with prolog or LISP or some language they think is AI related... lots of major attempts at developing large scale intelligence usually comes in to the limelight for a moment then disappears. It's VERY difficult since a computer really has no senses... no visual & audio input that is interpreted real-time. The only understanding a computer has of the world is what you tell it.

Intelligence would be the ability to classify objects and create new classifications. Understand language and context. Sort priorities according to where you are in society... it's hard to conceive... I think good AI won't be seen in our lifetime.
Jeff

Prolog and LISP are for setting up distinct rules that tell the computer what to do in case of an event. I am talking about truly evolutionary algorithms, so the language has little to do with it. I disagree that you need audio and visual input in real-time to be intelligent. Blind, deaf people are still intelligent.

Our brains are consturcted of neurons and firing synapses. It is complex, but when given any input, that input goes through our complex brain and comes to some sort of conclusion. By learning, we tweak these synapses. It works the same with evolutionary algorithms.

I implore you to read Blondie24. It is about real AI, not some scripted dictionary lookup like Deep Blue used to win at chess.

 

[complacent]

Originally posted by: jagec
The problem with AI is that it takes an insane amount of processing power. It just wouldn't work in a game...unless you wanted to give up on your pretty graphics.

The way computers work is they take solid facts, feed them through predetermined pathways, and get the result. This makes them lightening-fast. AI needs to work on fuzzy facts and preconcieved ideas acquired through experience, and constantly question whether they're using the right pathways or ideas to make a decision. Best to focus on our strengths: Use computers for computing, use humans for the fuzzy decisions.

The problem with thinking like this is you already "know" it can't be done. You are not the pioneering type and we are unlikely to read about any breakthroughs by people like you.

What is a fuzzy fact? Are you talking about fuzzy logic? Computers can be adapted to deal with fuzzy logic.

Our brain uses a neural network. A neural network can be constructed using a computer. For example, lets look at my mancala game. Essentially, you have two players each with 6 pits. The object of the game is to capture the most beads. The rules are extraneous and I will skip them.

In the nerual network, you could have 12 inputs, each representing the number of beads in each pit. You could also add inputs for current score, number of beads remaining, etc. etc. These values are sent through a vast neural network containing thousands of nodes and numbers, and after "seeing" ahead 6 moves or so (6 ply search), a decision was made to see what the best move would be. Essentially, I took a generation of 50 random networks and played a sample of games. The top 25 winners were copied, and the weights were varied randomly, and only by a little bit. Again, this was done only keeping the top 25 players, varying the weights, and playing again. Notice that the neural network had no knowledge of the game itself. It was only told whether it won or lost.

In the beginning, it was trivial to beat an opponent. By generation 10, it was somewhat more difficult, but I could still win. By generation 50, I could not win a single game against an evolved neural network. My sister-in-law, who is excellent at the game, only one 1 game out of 5.

Please, explain to me how this isn't adaptive learning.

 

[complacent]

Originally posted by: Jeffyboy
if you want examples... you could get a book called the "Little lisper"... which is about programming in LISP. (an AI language) You might find it a purdy cool language.

Jeff

I disagree. LISP is generally reffered to as an AI language...but it is a list processor. We are not talking about REAL artificial intelligence with LISP. LISP is used for setting up a list. You give it specific inputs, and it looks to see what the output should be.

Evolutionary programming is the way of the future. Read Blondie24 for more information.

 

[brand80]

wow....only 12 guesses on toilet paper

I am guessing that it is toilet paper?, Yes
12. Could it be found in a classroom? No.
11. Does it roll? Yes.
10. Does it fit in your wallet? No.
9. Can it be used more than once? No.
8. Does it help accomplish tasks? Yes.
7. Does it contain words? No.
6. Do you use it at night? Yes.
5. Is it made from paper? Yes.
4. Is it flat? Yes.
3. Do you clean it regularly? No.
2. Is it white? Yes.
1. It is classified as Other.


Uncommon Knowledge about toilet paper
Is it straight? I say No.
Is it used by the police? I say Yes.
Does it have a hole in it? I say Yes.
Does a hockey player use it? I say Yes.
Is it an insect? I say Probably.
Is it a pure-bred dog? I say Yes.
Does it have bumpy skin? I say Probably.
Does it spin? I say Yes.
Can it fit in an envelope? I say No.
Does it swim? I say Yes.
Is it a small mammal? I say Probably.
Can you find it in a church? I say Yes.
Does it live in groups(gregarious)? I say Probably.
Can you recycle it? I say Probably.
Can you sit on it? I say Yes.
Can it discharge waste? I say Probably.


LOL....still has some learning to do...

 

[cquark]

Originally posted by: complacent

Originally posted by: Jeffyboy
if you want examples... you could get a book called the "Little lisper"... which is about programming in LISP. (an AI language) You might find it a purdy cool language.

Jeff

I disagree. LISP is generally reffered to as an AI language...but it is a list processor. We are not talking about REAL artificial intelligence with LISP. LISP is used for setting up a list. You give it specific inputs, and it looks to see what the output should be.

You can do anything with LISP that you can with any standard programming language like C. LISP was designed for symbolic computing (symbolic mathematics, words, etc.) and follows the functional paradigm. C was designed for numerical computing and operating systems programming and follows the imperative programming paradigm.

Both are equally powerful in terms of what they can do. However, LISP's symbolic computing features and dynamic nature make it easier for programmers to write AI software in LISP than in C.

For the original poster, there are a number of books on AI programming for games these days. See
http://www.amazon.com/exec/obi...rm/102-3991510-9832960

 

[Jeff7]

Originally posted by: Jeffyboy
Problem with AI is that no one knows how to program AI. No one knows how the brain works either..lol.
Our attempts at AI include backward/forward chaining and goofy neural nets... it's all pretty primitive and made to sound impressive until you see the code that runs it. Give it a few more hundred years and I think we'll start to get closer ;-)

As for "fuzzy facts"... there's nothing really fuzzy in a computer... it's still 1's and 0's with probabilities associated to pathways.

Anyone have an algorigthm for parsing natural english language? It's freaking impossible almost. It's something that practically has to be learned... and i think we don't really know how to make computers learn things outside of what they are programmed to learn.

Jeff

Memory is another big problem. The brain uses very lossy compression, but it somehow figures out what is important enough to warrant being retained. You see a flock of birds outside the window, and you won't remember the number of birds, or what kind they are, but you might happen to remember the exact kind of airplane that was flying behind the birds. You see the birds first, but something about the plane makes your mind go "remember this information instead." Just the programming to make a computer know "this is important, I might need this," but to avoid having it do that for every bit of information it comes upon, would have to be very complex, at least by current computing standards.

Even if quantum computing happens in the next 100 years... it'll seriously be lacking programmability. In the brain... neurons are firing constantly and usually for no reason and usually nothing you may be aware of... bringing to your attention memories you aren't even trying to retrieve, and even memories you thought were 'deleted'. The brain is immensely complex to say the least.

It's like it's got random number generators all over the place. You can be sitting somewhere and just suddenly get a crazy mood swing and start thinking differently than before. What changed to make your mind alter its program set? Would a computer just be running Windows XP SP2, and then just say "Know what, I'm tired of this. Lets try an alpha build today." It'd be the computer equivalent of getting drunk.

 

[cquark]

/. just reviewed _AI for Computer Games_, and the book uses C++. See
http://books.slashdot.org/arti...3&tid=156&tid=6&tid=10