Energy Saving Additive Neural Network

TL;DR

This paper introduces an energy-efficient neural network architecture called additive neural network, which uses a novel ef-operator to eliminate multiplications, making it highly suitable for mobile devices while maintaining competitive performance.

Contribution

The paper proposes a new additive neural network with a multiplier-free ef-operator, enabling energy-efficient computations suitable for mobile computing.

Findings

Successfully solves the XOR problem.

Achieves classification performance similar to traditional neural networks on MNIST.

Uses a novel vector product based on ef-operator.

Abstract

In recent years, machine learning techniques based on neural networks for mobile computing become increasingly popular. Classical multi-layer neural networks require matrix multiplications at each stage. Multiplication operation is not an energy efficient operation and consequently it drains the battery of the mobile device. In this paper, we propose a new energy efficient neural network with the universal approximation property over space of Lebesgue integrable functions. This network, called, additive neural network, is very suitable for mobile computing. The neural structure is based on a novel vector product definition, called ef-operator, that permits a multiplier-free implementation. In ef-operation, the "product" of two real numbers is defined as the sum of their absolute values, with the sign determined by the sign of the product of the numbers. This "product" is used to construct a vector product in $R^N$. The vector product induces the $l_1$ norm. The proposed additive neural network successfully solves the XOR problem. The experiments on MNIST dataset show that the classification performances of the proposed additive neural networks are very similar to the corresponding multi-layer perceptron and convolutional neural networks (LeNet).