Jump to content

Image of the Day

BL00DG0R3
#screenshotsaturday #gamedev #indiedev https://t.co/SytiLEa4G5
IOTD | Top Screenshots

The latest, straight to your Inbox.

Subscribe to GameDev.net's newsletters to receive the latest updates and exclusive content.


Sign up now

ANN an alternative to evolving multiple instances

4: Adsense

Old topic!

Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.


  • You cannot reply to this topic
2 replies to this topic

#1 / rouncer   Members   

294
Like
0Likes
Like

Posted 10 February 2013 - 12:11 AM

Its fine that we can take 1000 instance generations and pick the most successful after a trial, but that has problems, what if you only have one instance? (say youd like it to learn off a user, not a program)

 

My idea, is maybe i can get an ANN to evolve in one instance.
 

Have a running success rate, that a mediator program modifies.

Add the success rate (a very small number), success can either be positive (reinforcing) or negative (inhibiting), from active synapses, only cells that fire get their output synapse permanences updated.   this rates the synapses for how generally successful they were.

then randomize the least successful synapse each frame, instead of the whole network.

 

So each frame, one synapse gets randomized.

 

Its a huge stab in the dark, what do you think?
 

 


Edited by rouncer, 10 February 2013 - 12:16 AM.


#2 alnite   Members   

3368
Like
0Likes
Like

Posted 10 February 2013 - 03:43 AM

A stab in the dark? Not really.  Sounds like backpropagation.



#3 Nickie   Members   

322
Like
1Likes
Like

Posted 10 February 2013 - 05:13 AM

There is one really good backprop tutorial. I needed 3 days to figure out what is happening, then I found it. Just from one read I implemented ANN library.

 

http://clemens.bytehammer.com/papers/BackProp/index.html

 

Good luck :)






Old topic!

Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.