The original paper of INLP uses a support vector machine and finds very similar results, because there isn’t actually a margin, data is always slightly mixed, but less when looking in the direction found by the linear classifier. (I implemented INLP with a linear classifier so that it could run on the GPU). I would be very surprised if it made any difference, given that L2 regularization on INLP doesn’t make a difference.
You could test this explanation using a support vector machine—it finds the direction that gives the maximum separation.
(This is a drive-by comment. I’m trying to reduce my external obligations, so I probably won’t be responding.)
The original paper of INLP uses a support vector machine and finds very similar results, because there isn’t actually a margin, data is always slightly mixed, but less when looking in the direction found by the linear classifier. (I implemented INLP with a linear classifier so that it could run on the GPU). I would be very surprised if it made any difference, given that L2 regularization on INLP doesn’t make a difference.