WPIA:accelerating DNN warm-up in Web browsers by precompiling WebGL programs  

作　　者：Deyu TIAN Yun MA Yudong HAN Qi YANG Haochen YANG Gang HUANG 

机构地区：[1]Key Laboratory of High Confidence Software Technologies(Peking University),Ministry of Education,School of Computer Science,Peking University,Beijing 100871,China [2]Institute for Artificial Intelligence,Peking University,Beijing 100871,China [3]National Key Laboratory of Data Space Technology and System,Beijing 100195,China

出　　处：《Frontiers of Computer Science》2024年第6期281-283,共3页计算机科学前沿（英文版）

基　　金：the National Natural Science Foundation of China(Grant No.62102009);the Beijing Outstanding Young Scientist Program(No.BJJWZYJH01201910001004)。

摘　　要：1 Introduction.With the advent of artificial intelligence,it is becoming increasingly common to leverage deep learning technologies to provide intelligent services in Web apps[1].Compared with traditional approaches where deep neural network(DNN)runs at cloud servers,DNN inference in Web browsers avoids serverconnections,boostingtheresponsivenesssand enhancing user privacy of Web apps[2].Typically,GPU acceleration is required during DNN inference,especially on enddevices.

关 键 词：WEB BROWSER WEBGL 

分 类 号：TP393.09[自动化与计算机技术—计算机应用技术] TP18[自动化与计算机技术—计算机科学与技术]