Alex, about your Counter Strike bot

Started by
4 comments, last by Tricron2 22 years, 2 months ago
I was reading a post where a user named Alex mentioned his project, a bot that used Neural Networks to play Counter-Strike (the half-life modification). It looked like you were doing the project for a university or organization or something. I was wondering if you had plans to release the bot publically on the internet when you were done. I think I would have many hours of fun trying to teach a bot to win counter-strike games. It would be like Pokemon, except fun and not gay.
Advertisement
Er... I think that''s me!

The bot is not CS (yet), I use Q2 for the moment. But, the AI is completely modular, and there are plans to port to Q3/UT/CS (time permitting).

Public releases are planned, yes! I want to make everything customiseable first, so people like you and I can have hours of fun Hopefully a first navigation demo will happen in a couple of weeks, so look for some activity about this.

Thanks for the interest!
Alex


Artificial Intelligence Depot - Maybe it''s not all about graphics...

Join us in Vienna for the nucl.ai Conference 2015, on July 20-22... Don't miss it!

Just wondering...How big is the brain??? I meen...How many inputs/outputs/levels? Would be fun to know

Hmm...Perhaps I should make the bots for our game work with NN''s...I guess it would take a long time to train them to be good bots :D
Lotsa playing with them..
delete this;
Another thing, if you made them half-life bots... It would be fun to observe a game full of players trained bots.
I''d train a bot, somebody else would train a bot, a few others, we would pit them into a battle to the death and the winning bots owner can brag...

Its kind of like a weird chatter program I once had. You trained the bots to ''try'' to talk (they did it poorly), and then you could hook up the too chatter bots and have them talk to each other.

It was kind of one dimensional. One bot kept swearing at the other, and the other kept complaining about Bill Clinton and the end of the world. Don''t ask.
Another thing, if you made them half-life bots... It would be fun to observe a game full of players trained bots.
I''d train a bot, somebody else would train a bot, a few others, we would pit them into a battle to the death and the winning bots owner can brag...

Its kind of like a weird chatter program I once had. You trained the bots to ''try'' to talk (they did it poorly), and then you could hook up the too chatter bots and have them talk to each other.

It was kind of one dimensional. One bot kept swearing at the other, and the other kept complaining about Bill Clinton and the end of the world. Don''t ask.
There''s not really one big brain. That doesn''t work well at all. There are quite a few NN, and lots of other AI algorithms too... modularity and flexibiliy is good, and a pure NN solution is ill-suited to that.

Indeed, I was planning on a ''bot wars'' as soon as the framework is robust and flexible enough... should be fun indeed!

Thanks for the support and ideas

Alex


Artificial Intelligence Depot - Maybe it''s not all about graphics...

Join us in Vienna for the nucl.ai Conference 2015, on July 20-22... Don't miss it!

This topic is closed to new replies.

Advertisement