Doctor of Crosswise: Reducing Over-parametrization in Neural Networks
Abstract
Dr. of Crosswise proposes a new architecture to reduce over-parametrization in Neural Networks. It introduces an operand for rapid computation in the framework of Deep Learning that leverages learned weights. The formalism is described in detail providing both an accurate elucidation of the mechanics and the theoretical implications.
Authors: J. de Curtò, I. de Zarzà, Kris Kitani, Irwin King, Michael Rung-Tsong Lyu
Published in: arXiv (Cornell University) (2019)