arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2110.04035cs.CV
收藏

UniNet:使用卷积、Transformer和MLP的统一架构搜索

UniNet: Unified Architecture Search with Convolution, Transformer, and MLP

CUHK-SenseTime Joint Laboratory(香港中文大学-商汤科技联合实验室);SenseTime Research(商汤科技研究院);The Chinese University of Hong Kong(香港中文大学)

Jihao Liu, Hongsheng Li, Guanglu Song, Xin Huang, Yu Liu

摘要:

最近,Transformer和多层感知器(MLP)架构在各种视觉任务上取得了令人印象深刻的结果。一些人工结合这些算子来设计可视化网络结构的研究工作,在一定程度上取得了令人满意的效果。在本文中,我们提出联合搜索卷积,Transformer,和MLP的最佳组合,以建立一系列的所有运营商的网络架构,具有高性能的视觉任务。我们的经验表明,广泛使用的跨卷积或池为基础的下采样模块成为性能瓶颈时,运营商组合形成一个网络。为了更好地处理由Transformer和MLP算子捕获的全局上下文,我们提出了两种新的上下文感知下采样模块,它们可以更好地适应由Transformer和MLP算子编码的全局信息。为此,我们在统一的搜索空间中联合搜索所有算子和下采样模块。值得注意的是,我们搜索到的网络UniNet(统一网络)在多个公共视觉基准测试、ImageNet分类、COCO目标检测和ADE 20 K语义分割方面的性能优于最先进的纯基于卷积的架构EfficientNet和纯基于transformer的架构Swin-Transformer。

英文摘要:

Recently, transformer and multi-layer perceptron (MLP) architectures have achieved impressive results on various vision tasks. A few works investigated manually combining those operators to design visual network architectures, and can achieve satisfactory performances to some extent. In this paper, we propose to jointly search the optimal combination of convolution, transformer, and MLP for building a series of all-operator network architectures with high performances on visual tasks. We empirically identify that the widely-used strided convolution or pooling based down-sampling modules become the performance bottlenecks when the operators are combined to form a network. To better tackle the global context captured by the transformer and MLP operators, we propose two novel context-aware down-sampling modules, which can better adapt to the global information encoded by transformer and MLP operators. To this end, we jointly search all operators and down-sampling modules in a unified search space. Notably, Our searched network UniNet (Unified Network) outperforms state-of-the-art pure convolution-based architecture, EfficientNet, and pure transformer-based architecture, Swin-Transformer, on multiple public visual benchmarks, ImageNet classification, COCO object detection, and ADE20K semantic segmentation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑