Modulated binary cliquenet

TL;DR

The paper introduces MBCliqueNet, a binarized CNN that significantly reduces storage requirements and achieves performance comparable to full-precision models like ResNet, enhancing portability for deployment on limited devices.

Contribution

It proposes a novel modulated binarized network with a new initialization method and parameter sharing, improving performance and reducing storage compared to existing binarized CNNs.

Findings

Reduces storage space by at least 32 times compared to full-precision models.

Achieves better performance than other binarized models.

Performs comparably or better than ResNet on tested datasets.

Abstract

Although Convolutional Neural Networks (CNNs) achieve effectiveness in various computer vision tasks, the significant requirement of storage of such networks hinders the deployment on computationally limited devices. In this paper, we propose a new compact and portable deep learning network named Modulated Binary Cliquenet (MBCliqueNet) aiming to improve the portability of CNNs based on binarized filters while achieving comparable performance with the full-precision CNNs like Resnet. In MBCliqueNet, we introduce a novel modulated operation to approximate the unbinarized filters and gives an initialization method to speed up its convergence. We reduce the extra parameters caused by modulated operation with parameters sharing. As a result, the proposed MBCliqueNet can reduce the required storage space of convolutional filters by a factor of at least 32, in contrast to the full-precision model, and achieve better performance than other state-of-the-art binarized models. More importantly, our model compares even better with some full-precision models like Resnet on the dataset we used.