Compounding the Performance Improvements of Assembled Techniques in a Convolutional Neural Network

TL;DR

This paper demonstrates that carefully assembling multiple CNN enhancement techniques can significantly improve accuracy and robustness while maintaining high throughput, validated across various tasks and datasets.

Contribution

The study introduces a method of combining existing CNN techniques to create a more accurate and robust model with minimal throughput loss, validated on multiple tasks.

Findings

Top-1 accuracy improved from 76.3% to 82.78%.

Achieved first place in CVPR 2019 iFood Competition.

Performance gains transferred effectively to other tasks.

Abstract

Recent studies in image classification have demonstrated a variety of techniques for improving the performance of Convolutional Neural Networks (CNNs). However, attempts to combine existing techniques to create a practical model are still uncommon. In this study, we carry out extensive experiments to validate that carefully assembling these techniques and applying them to basic CNN models (e.g. ResNet and MobileNet) can improve the accuracy and robustness of the models while minimizing the loss of throughput. Our proposed assembled ResNet-50 shows improvements in top-1 accuracy from 76.3\% to 82.78\%, mCE from 76.0\% to 48.9\% and mFR from 57.7\% to 32.3\% on ILSVRC2012 validation set. With these improvements, inference throughput only decreases from 536 to 312. To verify the performance improvement in transfer learning, fine grained classification and image retrieval tasks were tested on several public datasets and showed that the improvement to backbone network performance boosted transfer learning performance significantly. Our approach achieved 1st place in the iFood Competition Fine-Grained Visual Recognition at CVPR 2019, and the source code and trained models are available at https://github.com/clovaai/assembled-cnn