BATS: Binary ArchitecTure Search
This work addresses the problem of improving accuracy in binary neural networks for efficient deep learning applications, representing a novel method rather than an incremental improvement.
The paper tackles the accuracy gap between binary neural networks and real-valued networks by proposing BATS, a Neural Architecture Search framework tailored for binary networks, achieving new state-of-the-art results on CIFAR10, CIFAR100, and ImageNet datasets.
This paper proposes Binary ArchitecTure Search (BATS), a framework that drastically reduces the accuracy gap between binary neural networks and their real-valued counterparts by means of Neural Architecture Search (NAS). We show that directly applying NAS to the binary domain provides very poor results. To alleviate this, we describe, to our knowledge, for the first time, the 3 key ingredients for successfully applying NAS to the binary domain. Specifically, we (1) introduce and design a novel binary-oriented search space, (2) propose a new mechanism for controlling and stabilising the resulting searched topologies, (3) propose and validate a series of new search strategies for binary networks that lead to faster convergence and lower search times. Experimental results demonstrate the effectiveness of the proposed approach and the necessity of searching in the binary space directly. Moreover, (4) we set a new state-of-the-art for binary neural networks on CIFAR10, CIFAR100 and ImageNet datasets. Code will be made available https://github.com/1adrianb/binary-nas