LCNN: Lookup-based Convolutional Neural Network

11/20/2016
by   Hessam Bagherinezhad, et al.
0

Porting state of the art deep learning algorithms to resource constrained compute platforms (e.g. VR, AR, wearables) is extremely challenging. We propose a fast, compact, and accurate model for convolutional neural networks that enables efficient learning and inference. We introduce LCNN, a lookup-based convolutional neural network that encodes convolutions by few lookups to a dictionary that is trained to cover the space of weights in CNNs. Training LCNN involves jointly learning a dictionary and a small set of linear combinations. The size of the dictionary naturally traces a spectrum of trade-offs between efficiency and accuracy. Our experimental results on ImageNet challenge show that LCNN can offer 3.2x speedup while achieving 55.1 AlexNet architecture. Our fastest LCNN offers 37.6x speed up over AlexNet while maintaining 44.3 inference, but it also enables efficient training. In this paper, we show the benefits of LCNN in few-shot learning and few-iteration learning, two crucial aspects of on-device training of deep learning models.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
05/21/2018

AgileNet: Lightweight Dictionary-based Few-shot Learning

The success of deep learning models is heavily tied to the use of massiv...
research
07/07/2015

Dependency-based Convolutional Neural Networks for Sentence Embedding

In sentence modeling and classification, convolutional neural network ap...
research
04/14/2023

Real Time Bearing Fault Diagnosis Based on Convolutional Neural Network and STM32 Microcontroller

With the rapid development of big data and edge computing, many research...
research
09/19/2018

Characterising Across-Stack Optimisations for Deep Convolutional Neural Networks

Convolutional Neural Networks (CNNs) are extremely computationally deman...
research
10/08/2021

LCS: Learning Compressible Subspaces for Adaptive Network Compression at Inference Time

When deploying deep learning models to a device, it is traditionally ass...
research
08/25/2020

JokeMeter at SemEval-2020 Task 7: Convolutional humor

This paper describes our system that was designed for Humor evaluation w...
research
09/04/2020

ACDC: Weight Sharing in Atom-Coefficient Decomposed Convolution

Convolutional Neural Networks (CNNs) are known to be significantly over-...

Please sign up or login with your details

Forgot password? Click here to reset