DVOLVER: Efficient Pareto-Optimal Neural Network Architecture Search

02/05/2019
by   Guillaume Michel, et al.
0

Automatic search of neural network architectures is a standing research topic. In addition to the fact that it presents a faster alternative to hand-designed architectures, it can improve their efficiency and for instance generate Convolutional Neural Networks (CNN) adapted for mobile devices. In this paper, we present a multi-objective neural architecture search method to find a family of CNN models with the best accuracy and computational resources tradeoffs, in a search space inspired by the state-of-the-art findings in neural search. Our work, called Dvolver, evolves a population of architectures and iteratively improves an approximation of the optimal Pareto front. Applying Dvolver on the model accuracy and on the number of floating points operations as objective functions, we are able to find, in only 2.5 days, a set of competitive mobile models on ImageNet. Amongst these models one architecture has the same Top-1 accuracy on ImageNet as NASNet-A mobile with 8 floating point operations and another one has a Top-1 accuracy of 75.28 ImageNet exceeding by 0.28 computational resources.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
04/24/2018

Multi-objective Architecture Search for CNNs

Architecture search aims at automatically finding neural architectures t...
research
04/08/2019

Resource Constrained Neural Network Architecture Search

The design of neural network architectures is frequently either based on...
research
09/24/2019

Constrained deep neural network architecture search for IoT devices accounting hardware calibration

Deep neural networks achieve outstanding results in challenging image cl...
research
07/23/2020

PareCO: Pareto-aware Channel Optimization for Slimmable Neural Networks

Slimmable neural networks provide a flexible trade-off front between pre...
research
09/26/2019

Exascale Deep Learning to Accelerate Cancer Research

Deep learning, through the use of neural networks, has demonstrated rema...
research
04/08/2020

GeneCAI: Genetic Evolution for Acquiring Compact AI

In the contemporary big data realm, Deep Neural Networks (DNNs) are evol...
research
05/06/2019

Fast and Reliable Architecture Selection for Convolutional Neural Networks

The performance of a Convolutional Neural Network (CNN) depends on its h...

Please sign up or login with your details

Forgot password? Click here to reset