Much to specialized for most people but anyway you can experiment a little if like:
"I was having a problem evolving deep networks where the information loss caused by the dot product weight and sum operations was greater than the computational gains per layer.
By allowing more ways for information to pass through each layer I am seeing markedly improved results.
https://drive.google.com/open?id=0BwsgM ... DFCcjJnQTA
or
https://drive.google.com/open?id=0BwsgM ... UlreWZSVXc
"
Discussion:
https://software.intel.com/en-us/forums ... pic/734095