Practical Block-Wise Neural Network Architecture Generation

Zhao Zhong, Junjie Yan, Wei Wu, Jing Shao, Cheng-Lin Liu; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018, pp. 2423-2432


Convolutional neural networks have gained a remarkable success in computer vision. However, most usable network architectures are hand-crafted and usually require expertise and elaborate design. In this paper, we provide a block-wise network generation pipeline called BlockQNN which automatically builds high-performance networks using the Q-Learning paradigm with epsilon-greedy exploration strategy. The optimal network block is constructed by the learning agent which is trained sequentially to choose component layers. We stack the block to construct the whole auto-generated network. To accelerate the generation process, we also propose a distributed asynchronous framework and an early stop strategy.The block-wise generation brings unique advantages: (1) it performs competitive results in comparison to the hand-crafted state-of-the-art networks on image classification, additionally, the best network generated by BlockQNN achieves 3.54% top-1 error rate on CIFAR-10 which beats all existing auto-generate networks. (2) in the meanwhile, it offers tremendous reduction of the search space in designing networks which only spends 3 days with 32 GPUs, and (3) moreover, it has strong generalizability that the network built on CIFAR also performs well on a larger-scale ImageNet dataset.

Related Material

[pdf] [Supp] [arXiv]
author = {Zhong, Zhao and Yan, Junjie and Wu, Wei and Shao, Jing and Liu, Cheng-Lin},
title = {Practical Block-Wise Neural Network Architecture Generation},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2018}