Closed floybix closed 9 years ago
Within a layer there are two types of active state that cells can be in (not including predictive):
Both states count towards depolarisation of other cells, and reinforcing synapses, but only the former is used when growing new synapses. I imagine the distinction biologically would be one of timing and duration of firing.
To generalise the current within-layer method across layers, we will need to pass both active states across layers. I'm referring to them as learn-cells (1) and active-cells (1 + 2).
On the other hand, it's not clear what bursting in a motor command would mean, and I'm not sure biologically that the distinction would carry across layers if there are intermediary mechanisms. Leaving it for now.
done in 7bfe1e9
Currently distal synapses refer to a source cell in the same layer. Need to allow the source to be an input bit either from below (motor) or above (feedback) in the hierarchy.
So the source set will become a set of integers which can be mapped back to local cells and/or input bits.
I'm working on this now.