Training Efficient CNNS: Tweaking the Nuts and Bolts of Neural Networks for Lighter, Faster and Robust Models

05/23/2022
by   Sabeesh Ethiraj, et al.
0

Deep Learning has revolutionized the fields of computer vision, natural language understanding, speech recognition, information retrieval and more. Many techniques have evolved over the past decade that made models lighter, faster, and robust with better generalization. However, many deep learning practitioners persist with pre-trained models and architectures trained mostly on standard datasets such as Imagenet, MS-COCO, IMDB-Wiki Dataset, and Kinetics-700 and are either hesitant or unaware of redesigning the architecture from scratch that will lead to better performance. This scenario leads to inefficient models that are not suitable on various devices such as mobile, edge, and fog. In addition, these conventional training methods are of concern as they consume a lot of computing power. In this paper, we revisit various SOTA techniques that deal with architecture efficiency (Global Average Pooling, depth-wise convolutions squeeze and excitation, Blurpool), learning rate (Cyclical Learning Rate), data augmentation (Mixup, Cutout), label manipulation (label smoothing), weight space manipulation (stochastic weight averaging), and optimizer (sharpness aware minimization). We demonstrate how an efficient deep convolution network can be built in a phased manner by sequentially reducing the number of training parameters and using the techniques mentioned above. We achieved a SOTA accuracy of 99.2 an accuracy of 86.01

READ FULL TEXT

page 1

page 2

page 3

page 4

research
06/16/2021

Efficient Deep Learning: A Survey on Making Deep Learning Models Smaller, Faster, and Better

Deep Learning has revolutionized the fields of computer vision, natural ...
research
07/15/2020

Data-Efficient Deep Learning Method for Image Classification Using Data Augmentation, Focal Cosine Loss, and Ensemble

In general, sufficient data is essential for the better performance and ...
research
04/23/2019

DenseNet Models for Tiny ImageNet Classification

In this paper, we present two image classification models on the Tiny Im...
research
08/08/2022

Efficient Neural Net Approaches in Metal Casting Defect Detection

One of the most pressing challenges prevalent in the steel manufacturing...
research
04/27/2016

Diving deeper into mentee networks

Modern computer vision is all about the possession of powerful image rep...
research
05/14/2022

Efficient Deep Learning Methods for Identification of Defective Casting Products

Quality inspection has become crucial in any large-scale manufacturing i...

Please sign up or login with your details

Forgot password? Click here to reset