论文标题
BIMLP：视觉多层感知的紧凑型二元架构
BiMLP: Compact Binary Architectures for Vision Multi-Layer Perceptrons
论文作者
论文摘要
本文研究了为视觉多层感知器（MLP）设计紧凑的二元架构的问题。我们提供了有关二进制视力MLP难度的广泛分析，并发现由于二进制MLP的能力有限，以前的二进制方法的性能很差。与使用具有较大内核大小的卷积操作的传统CNN相比，MLP中的完全连接（FC）层可以视为内核尺寸$ 1 \ times1 $的卷积层。因此，FC层的表示能力在被二进制时将受到限制，并且对空间混合和通道混合能力的限制在中间特征上。为此，我们建议通过丰富二进制FC层的表示能力来提高二进制MLP（BIMLP）模型的性能。我们设计了一个新型的二进制块，其中包含多个分支，以合并同一阶段的一系列输出，也是一种通用的快捷连接，鼓励从上一个阶段开始信息流。还要仔细设计下采样层，以降低计算复杂性，同时保持分类性能。基准数据集Imagenet-1K上的实验结果证明了所提出的BIMLP模型的有效性，该模型与先前的二进制CNN相比达到了最先进的精度。 Mindspore代码可在\ url {https://gitee.com/mindspore/models/tree/master/master/research/cv/bimlp}中获得。
This paper studies the problem of designing compact binary architectures for vision multi-layer perceptrons (MLPs). We provide extensive analysis on the difficulty of binarizing vision MLPs and find that previous binarization methods perform poorly due to limited capacity of binary MLPs. In contrast with the traditional CNNs that utilizing convolutional operations with large kernel size, fully-connected (FC) layers in MLPs can be treated as convolutional layers with kernel size $1\times1$. Thus, the representation ability of the FC layers will be limited when being binarized, and places restrictions on the capability of spatial mixing and channel mixing on the intermediate features. To this end, we propose to improve the performance of binary MLP (BiMLP) model by enriching the representation ability of binary FC layers. We design a novel binary block that contains multiple branches to merge a series of outputs from the same stage, and also a universal shortcut connection that encourages the information flow from the previous stage. The downsampling layers are also carefully designed to reduce the computational complexity while maintaining the classification performance. Experimental results on benchmark dataset ImageNet-1k demonstrate the effectiveness of the proposed BiMLP models, which achieve state-of-the-art accuracy compared to prior binary CNNs. The MindSpore code is available at \url{https://gitee.com/mindspore/models/tree/master/research/cv/BiMLP}.
