News for hackers. Stuff that matters.
Re:Deep Quaternion Networks (2017) [pdf] by gaudetcj on Friday March 23, 2018 @08:55AM (#16657528)
Hi I'm one of the authors of this paper. Sorry if it unclear, but we reduce the number of filters per layer to account for this. The quaternion networks actually have fewer parameters.
Re:Deep Quaternion Networks (2017) [pdf] by highd on Saturday March 24, 2018 @01:49AM (#16665083)
Can you provide some technical details on what you do? Do you divide the number of channels on each real layered network by 4? I don't see anything describing this in the paper.
Re:Deep Quaternion Networks (2017) [pdf] by gaudetcj on Tuesday March 27, 2018 @09:35AM (#16688148)
Yes that is exactly what we do.