Jump to content
  • Advertisement

Archived

This topic is now archived and is closed to further replies.

pacman2003

WOW YOU JUST GOTTA SEE THIS!!

This topic is 5520 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

Advertisement
yeah i have to agree: very cool indeed. the first functional NeuralNet ive ever seen!
its behaviour is really very interesting. the firsy time i ran it, the thing had some very negative experiences when putting its foot down, cos it happended to move its 'toes' the wrong way, so it decided it shouldnt thouch the ground anymore altogether. the second time it learned really well though, it moves forward now with quite a decent speed .
also funy that it learns some bad habits. my simulation has a habit of pecking the ground a few times after a few succesfull moves from the beginning, and it doesnt seem to get rid of it.

EDIT: is it a NN btw... i assumed it but it is never stated anywhere, it just is said that it learns..

[edited by - Eelco on May 3, 2003 3:38:48 PM]

Share this post


Link to post
Share on other sites
Going on 5 minutes.. and my guy isn't moving. He's not the brightest AI i've ever seen. Matter of fact, he's not even close to moving.


--- Edit ---
I am experiencing the same problem as the guy above . I guess it's time to restart/refresh.

[edited by - Ready4Dis on May 3, 2003 3:37:12 PM]

Share this post


Link to post
Share on other sites
Guest Anonymous Poster
just skip a few thousands of steps and you''ll see it doing the 100 meter in 1 second .

Share this post


Link to post
Share on other sites
Cool! I just finished my first experiment into NeuralNets, and I come online to find this (Mine arnt as cool (they learn very slowly)).

I think my next project is gonna be a basic physics sim to stick my neural nets in

George. F
"Who says computer games affect kids, imagine if PacMan affected us as kids, we''d all sit around in a darkened room munching pills and listening to repetitive music....uh oh!"

Share this post


Link to post
Share on other sites
Are you all sure this is a neural net? Read the paper on the robot which can be found at robot paper , the 5th entry in Papers .

Eric


[edited by - Geta on May 4, 2003 10:04:37 AM]

Share this post


Link to post
Share on other sites
hi,

I think they use reinforcement learning and partially observable markov decision processes (POMDPs). That''s quite different to neural nets.

/Mankind gave birth to God.

Share this post


Link to post
Share on other sites
Guest Anonymous Poster
**great**

Share this post


Link to post
Share on other sites

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

Participate in the game development conversation and more when you create an account on GameDev.net!

Sign me up!