Optimizing Convolutional Neural Network Architecture via Information Field

by   Yuke Wang, et al.

CNN architecture design has attracted tremendous attention of improving model accuracy or reducing model complexity. However, existing works either introduce repeated training overhead in the search process or lack an interpretable metric to guide the design. To clear the hurdles, we propose Information Field (IF), an explainable and easy-to-compute metric, to estimate the quality of a CNN architecture and guide the search process of designs. To validate the effectiveness of IF, we build a static optimizer to improve the CNN architectures at both the stage level and the kernel level. Our optimizer not only provides a clear and reproducible procedure but also mitigates unnecessary training efforts in the architecture search process. Experiments show that the models generated by our optimizer can achieve up to 5.47 and up to 65.38 structures like MobileNet and ResNet.



There are no comments yet.


page 4

page 5


S2DNAS: Transforming Static CNN Model for Dynamic Inference via Neural Architecture Search

Recently, dynamic inference has emerged as a promising way to reduce the...

GP-CNAS: Convolutional Neural Network Architecture Search with Genetic Programming

Convolutional neural networks (CNNs) are effective at solving difficult ...

ENAS4D: Efficient Multi-stage CNN Architecture Search for Dynamic Inference

Dynamic inference is a feasible way to reduce the computational cost of ...

AutoSpeech: Neural Architecture Search for Speaker Recognition

Speaker recognition systems based on Convolutional Neural Networks (CNNs...

ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design

Currently, the neural network architecture design is mostly guided by th...

ACNet: Strengthening the Kernel Skeletons for Powerful CNN via Asymmetric Convolution Blocks

As designing appropriate Convolutional Neural Network (CNN) architecture...

The Nonlinearity Coefficient - A Practical Guide to Neural Architecture Design

In essence, a neural network is an arbitrary differentiable, parametrize...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.