Advertisement Jump to content


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


Feedforward Networks Information

This topic is 5478 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

hi, does anyone know anywhere that i can find accurate information about "Feedforward Networks with perturbation based learning"... ive been looking all over the internet but i cant find much about it..i belive its where you add a value to the weight along the way, a man named John Manslow demonstrated this in a tanks game... but as i said i havent found much else out about this..any help would be great thanks...

Share this post

Link to post
Share on other sites

Although they can be quite effective, perturbation searches are only very rarely used to train neural networks. I used them in the GPG2 gem because it''s easy to understand how they work without knowing much about neural networks or having a strong maths background. After all, the gem was designed only as an introduction.

The type of neural network used in the gem was a mutlilayer perceptron (MLP), which are most commonly trained using backpropagation gradient descent. You should find quite a lot of information about that on the internet if you don''t like the perturbation search.



Share this post

Link to post
Share on other sites

  • Advertisement

Important Information

By using, you agree to our community Guidelines, Terms of Use, and Privacy Policy. is your game development community. Create an account for your GameDev Portfolio and participate in the largest developer community in the games industry.

Sign me up!