Knowledge Commons of Institute of Automation,CAS
The Devil is in Details: Delving Into Lite FFN Design for Vision Transformers | |
Chen, Zhiyang1,2![]() ![]() ![]() ![]() ![]() ![]() | |
2024-03-18 | |
会议名称 | 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) |
会议日期 | 2024-4-14 |
会议地点 | Seoul, Korea |
摘要 | Transformer has demonstrated exceptional performance on a variety of vision tasks. However, its high computational complexity can become problematic. In this paper, we conduct a systematic analysis of the complexity of each component in vision transformers, and identify an easily overlooked detail: that the Feed-Forward Network (FFN) is the primary computational bottleneck, even more so than the Multi-Head Self-Attention (MHSA) mechanism. Inspired by this, we further propose a lightweight FFN module, named SparseFFN, that can reduce dense computations in both channel and spatial dimension. Specifically, SparseFFN consists of two components: Channel-Sparse FFN (CS-FFN) and Spatial-Sparse FFN (SS-FFN), which can be seamlessly incorporated into various vision transformers and even pure MLP models with significantly fewer FLOPs. Extensive experiments demonstrate the effectiveness and efficiency of the proposed method. For example, our approach can reduce model complexity by 23%-39% for most of vision transformers and MLP models while keeping comparable accuracy. |
关键词 | Vision Transformer Light-Weight Structure Feed-Forward Networks |
收录类别 | EI |
七大方向——子方向分类 | 图像视频处理与分析 |
国重实验室规划方向分类 | 视觉信息处理 |
是否有论文关联数据集需要存交 | 否 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/56594 |
专题 | 紫东太初大模型研究中心_大模型计算 |
作者单位 | 1.Foundation Model Research Center, Institute of Automation, Chinese Academy of Sciences 2.School of Artificial Intelligence, University of Chinese Academy of Sciences 3.Peng Cheng Laboratory 4.Wuhan AI Research |
第一作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Chen, Zhiyang,Zhu, Yousong,Li, Zhaowen,et al. The Devil is in Details: Delving Into Lite FFN Design for Vision Transformers[C],2024. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
ICASSP2024 (1).pdf(407KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论