Fixed-point Quantization for Vision Transformer | |
Zhexin, Li2,3; Peisong Wang3; Zhiyuan Wang1; Jian Cheng3 | |
2022-03-14 | |
会议日期 | 2021-10-22 |
会议地点 | Beijing, China |
DOI | 10.1109/CAC53003.2021.9728246 |
英文摘要 | Recently, transformer-based models has shown promising results on miscellaneous computer vision tasks. However, its high computation cost makes it neither practical to deploy on mobile devices, nor economic to compute on servers. In this paper, we propose two effective quantization schemes for reducing the memory usage and computation consumption of vision transformers. First, we develop an approximation-based Post-training Quantization (PTQ) approach which optimizes for a set of quantization scaling factors that minimize quantization errors. Moreover, we introduce a learning-based Quantization-aware Training (QAT) approach that allows for model finetuning after inserting quantization operations to restore accuracy. Furthermore, we reveal the complementary effects of learning-based approach and approximation-based approach in QAT and propose an effective strategy for the initialization of quantization parameters. We evaluate our approaches on ImageNet for different vision transformer models. Our quantization algorithms outperform the previous state-of-art approaches for both post-training quantization and quantization-aware training benchmark. With weights and activations in vision transformer quantized to 8-bit integers, we obtain a ×4 compression rate of model parameters with an accuracy drop of less than 0.2% for models of various scales. |
内容类型 | 会议论文 |
源URL | [http://ir.ia.ac.cn/handle/173211/48703] |
专题 | 类脑芯片与系统研究 |
通讯作者 | Zhiyuan Wang |
作者单位 | 1.Artificial Intelligence Research Center (AIRC), Defense Innovation Institute (DII), Beijing, China 2.School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China 3.NLPR & AiRiA, Institute of Automation, Chinese Academy of Sciences, Beijing, China |
推荐引用方式 GB/T 7714 | Zhexin, Li,Peisong Wang,Zhiyuan Wang,et al. Fixed-point Quantization for Vision Transformer[C]. 见:. Beijing, China. 2021-10-22. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
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