MINCONVNETS: A NEW CLASS OF MULTIPLICATION-LESS NEURAL NETWORKS

In this article, MinConvNets where the multiplications in the forward propagation path of CNNs are approximated by minimum comparator operations are introduced. Hardware complexity of minimum operator is of the order of O(N), whereas for multiplication it is O(N²). Firstly, a methodology to find approximate operations based on statistical correlation is presented. We show that it is possible to replace multipliers by minimum operations in the forward propagation under certain constraints, i.e. given similar mean and variances of the feature and the weight vectors. A modified training method which guarantees the above constraints is proposed. And it is shown that equivalent precision can be achieved during inference with MinConvNets by using transfer learning from well trained exact CNNs.