Convolutional Neural Networks at Constrained Time Cost

TL;DR

This paper explores how to design CNN architectures that balance accuracy and computational time, demonstrating a model that achieves high accuracy with 20% less inference time than AlexNet.

Contribution

It introduces a method for modifying CNN architectures under time constraints, providing insights into the trade-offs among design factors and presenting a competitive, faster model.

Findings

Achieved 11.8% top-5 error on ImageNet with a 20% faster model than AlexNet.

Demonstrated the importance of architectural factors like depth and filter size under time constraints.

Provided controlled comparisons to understand the impact of design choices.

Abstract

Though recent advanced convolutional neural networks (CNNs) have been improving the image recognition accuracy, the models are getting more complex and time-consuming. For real-world applications in industrial and commercial scenarios, engineers and developers are often faced with the requirement of constrained time budget. In this paper, we investigate the accuracy of CNNs under constrained time cost. Under this constraint, the designs of the network architectures should exhibit as trade-offs among the factors like depth, numbers of filters, filter sizes, etc. With a series of controlled comparisons, we progressively modify a baseline model while preserving its time complexity. This is also helpful for understanding the importance of the factors in network designs. We present an architecture that achieves very competitive accuracy in the ImageNet dataset (11.8% top-5 error, 10-view test), yet is 20% faster than "AlexNet" (16.0% top-5 error, 10-view test).