EuclidNets: An Alternative Operation for Efficient Inference of Deep Learning Models

TL;DR

EuclidNet introduces a novel operation replacing multiplication with Euclidean distance to enable more efficient deep learning inference on edge devices, maintaining performance while reducing computational complexity.

Contribution

The paper proposes EuclidNet, a new operation for deep learning models that replaces multiplication with Euclidean distance, facilitating hardware-efficient implementations.

Findings

EuclidNet aligns with matrix multiplication for convolutional layers.

It maintains comparable performance under various transformations and noise.

It enables more efficient hardware implementation for edge devices.

Abstract

With the advent of deep learning application on edge devices, researchers actively try to optimize their deployments on low-power and restricted memory devices. There are established compression method such as quantization, pruning, and architecture search that leverage commodity hardware. Apart from conventional compression algorithms, one may redesign the operations of deep learning models that lead to more efficient implementation. To this end, we propose EuclidNet, a compression method, designed to be implemented on hardware which replaces multiplication, $xw$, with Euclidean distance $(x-w)^2$. We show that EuclidNet is aligned with matrix multiplication and it can be used as a measure of similarity in case of convolutional layers. Furthermore, we show that under various transformations and noise scenarios, EuclidNet exhibits the same performance compared to the deep learning models designed with multiplication operations.