bug with IAC?

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

bug with IAC?

Gaurav Suri
If I create 2 pairs of nodes -- call them A, B, C and D

Say I connect A with B (bidirectional, excitatory) and C with D (also bidirectional, excitatory). A and B are not connected to C and D.

Now if I activate A to 1, one would expect B to also activate to 1. This happens. Unfortunately after enough iterations, C and D also activate to 1. This should not happen.

Sorry if I'm doing something silly; or is this a bug?
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: bug with IAC?

Gaurav Suri
My bad, this is no bug. The IAC neurons start at 0.1 activation. Connecting 2 such neurons to each other ensures that they will reach max activation without any external input.

Sorry!
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: bug with IAC?

jyoshimi
Administrator
In reply to this post by Gaurav Suri
Hi there,

This is because IAC neurons have a .01 resting value by default.  One IAC neuron will rise to .01 after a while.  If you bidirectionally connect them this will max them both out pretty quick.

Thanks for the note!

- jeff

Loading...