Nice: Noise injection and clamping estimation for neural network quantization

Chaim Baskin, Evgenii Zheltonozhkii, Tal Rozen, Natan Liss, Yoav Chai, Eli Schwartz, Raja Giryes, Alexander M. Bronstein, Avi Mendelson

Research output: Contribution to journalArticlepeer-review

Abstract

Convolutional Neural Networks (CNNs) are very popular in many fields including computer vision, speech recognition, natural language processing, etc. Though deep learning leads to groundbreaking performance in those domains, the networks used are very computationally de-manding and are far from being able to perform in real-time applications even on a GPU, which is not power efficient and therefore does not suit low power systems such as mobile devices. To overcome this challenge, some solutions have been proposed for quantizing the weights and activations of these networks, which accelerate the runtime significantly. Yet, this acceleration comes at the cost of a larger error unless spatial adjustments are carried out. The method proposed in this work trains quantized neural networks by noise injection and a learned clamping, which improve accuracy. This leads to state-of-the-art results on various regression and classification tasks, e.g., ImageNet classification with architectures such as ResNet-18/34/50 with as low as 3 bit weights and activations. We implement the proposed solution on an FPGA to demonstrate its applicability for low-power real-time applications. The quantization code will become publicly available upon acceptance.

Original languageEnglish
Article number2144
JournalMathematics
Volume9
Issue number17
DOIs
StatePublished - Sep 2021

Keywords

  • CNN architecture
  • Low power
  • Neural networks
  • Quantization

Fingerprint

Dive into the research topics of 'Nice: Noise injection and clamping estimation for neural network quantization'. Together they form a unique fingerprint.

Cite this