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 2304.11520cs.CLcs.SYeess.SY
收藏

嵌入式设备上的自然语言处理:Transformer模型的表现如何?

Processing Natural Language on Embedded Devices: How Well Do Transformer Models Perform?

Auburn University(奥本大学);Washington State University(华盛顿州立大学)

Souvika Sarkar, Mohammad Fakhruddin Babar, Md Mahadi Hassan, Monowar Hasan, Shubhra Kanti Karmaker Santu

AI总结:

该研究测试了多款基于BERT的Transformer模型在不同配置嵌入式设备上的多任务性能,验证了无GPU的低配置嵌入式设备也能运行复杂NLP任务,为模型嵌入式部署提供参考。

摘要:

本文针对不同硬件配置和精度要求下的Transformer语言模型开展性能研究,得出关于资源与精度权衡的实证结论。我们重点研究了最常用的基于BERT的语言模型(即BERT、RoBERTa、DistilBERT和TinyBERT)在嵌入式系统上的表现。我们在4款现成嵌入式平台(Raspberry Pi、Jetson、UP2和UDOO)的2GB和4GB内存配置(共8种硬件配置)上,搭配4个数据集(即HuRIC、GoEmotion、CoNLL、WNUT17)运行各类自然语言处理任务进行测试。研究发现,即便无GPU(例如配备2GB RAM的Raspberry Pi),在嵌入式系统上执行复杂自然语言处理任务(如“情感”分类)也具备可行性。研究结果可帮助设计者理解Transformer语言模型、尤其是基于BERT架构的模型的可部署性与性能表现。

英文摘要:

This paper presents a performance study of transformer language models under different hardware configurations and accuracy requirements and derives empirical observations about these resource/accuracy trade-offs. In particular, we study how the most commonly used BERT-based language models (viz., BERT, RoBERTa, DistilBERT, and TinyBERT) perform on embedded systems. We tested them on four off-the-shelf embedded platforms (Raspberry Pi, Jetson, UP2, and UDOO) with 2 GB and 4 GB memory (i.e., a total of eight hardware configurations) and four datasets (i.e., HuRIC, GoEmotion, CoNLL, WNUT17) running various NLP tasks. Our study finds that executing complex NLP tasks (such as "sentiment" classification) on embedded systems is feasible even without any GPUs (e.g., Raspberry Pi with 2 GB of RAM). Our findings can help designers understand the deployability and performance of transformer language models, especially those based on BERT architectures.

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。

↑