GBCNs: Genetic Binary Convolutional Networks for Enhancing the Performance of 1-bit DCNNs
This work addresses training inefficiency and accuracy loss in 1-bit DCNNs for computer vision applications, representing an incremental improvement with a novel method for a known bottleneck.
The paper tackles the problem of training 1-bit deep convolutional neural networks (DCNNs), which often get trapped in local minima due to binarized kernels and activations, by proposing Genetic Binary Convolutional Networks (GBCNs) with a balanced Genetic Algorithm (BGA) to improve performance, achieving strong generalization on tasks like CIFAR, ImageNet, face recognition, and re-identification.
Training 1-bit deep convolutional neural networks (DCNNs) is one of the most challenging problems in computer vision, because it is much easier to get trapped into local minima than conventional DCNNs. The reason lies in that the binarized kernels and activations of 1-bit DCNNs cause a significant accuracy loss and training inefficiency. To address this problem, we propose Genetic Binary Convolutional Networks (GBCNs) to optimize 1-bit DCNNs, by introducing a new balanced Genetic Algorithm (BGA) to improve the representational ability in an end-to-end framework. The BGA method is proposed to modify the binary process of GBCNs to alleviate the local minima problem, which can significantly improve the performance of 1-bit DCNNs. We develop a new BGA module that is generic and flexible, and can be easily incorporated into existing DCNNs, such asWideResNets and ResNets. Extensive experiments on the object classification tasks (CIFAR, ImageNet) validate the effectiveness of the proposed method. To highlight, our method shows strong generalization on the object recognition task, i.e., face recognition, facial and person re-identification.