Pyramid Vector Quantization for Deep Learning

TL;DR

This paper introduces Pyramid Vector Quantization (PVQ) as a method to significantly reduce computational costs and compress neural network weights by enabling dot product calculations with only additions, subtractions, and one multiplication.

Contribution

The paper presents PVQ as a novel quantization technique that simplifies neural network computations and compression, applicable to any architecture based on dot products.

Findings

PVQ enables dot product computation with minimal operations.

Neural networks can be compressed and accelerated using PVQ.

Applicable to various neural network architectures.

Abstract

This paper explores the use of Pyramid Vector Quantization (PVQ) to reduce the computational cost for a variety of neural networks (NNs) while, at the same time, compressing the weights that describe them. This is based on the fact that the dot product between an N dimensional vector of real numbers and an N dimensional PVQ vector can be calculated with only additions and subtractions and one multiplication. This is advantageous since tensor products, commonly used in NNs, can be re-conduced to a dot product or a set of dot products. Finally, it is stressed that any NN architecture that is based on an operation that can be re-conduced to a dot product can benefit from the techniques described here.