On the cost effectiveness of logarithmic arithmetic for backpropagation training on SIMD processors
Mark G. Arnold, T. A. Bailey, Jerry J. Cupal, M.D. Winkel · Proceedings of International Conference on Neural Networks (ICNN'97) · 2002
We show that backpropagation on a SIMD processor with logarithmic arithmetic uses less memory and may be up to 3.2 times more cost effective than on one with fixed point arithmetic when synthesized in the same technology.