2024暑期iHUB·上海：計(jì)算機(jī)科學(xué) 云存儲(chǔ)與云計(jì)算專題：數(shù)據(jù)存儲(chǔ)的算法原理及其在分布式人工智能系統(tǒng)與區(qū)塊鏈系統(tǒng)中的應(yīng)用研究【大學(xué)組】

專業(yè)：人工智能,計(jì)算機(jī)科學(xué),工程

項(xiàng)目類型：海外導(dǎo)師線下項(xiàng)目

開始時(shí)間：2024年07月20日

是否可加論文：是

項(xiàng)目周期：1周在線科研+14天面授科研+5周在線論文指導(dǎo)

語言：英文

有無剩余名額：名額充足

建議學(xué)生年級(jí)：大學(xué)生

是否必需面試：否

適合專業(yè)：計(jì)算機(jī)科學(xué)物聯(lián)網(wǎng)網(wǎng)絡(luò)與信息安全電子工程軟件工程信號(hào)與信息處理數(shù)據(jù)科學(xué)深度學(xué)習(xí)數(shù)據(jù)結(jié)構(gòu)與算法

地點(diǎn)：上海圣華紫竹學(xué)院

建議選修：EEE電子與電氣工程核心：信號(hào)與系統(tǒng)

建議具備的基礎(chǔ)：計(jì)算機(jī)科學(xué)、軟件工程、網(wǎng)絡(luò)與信息安全相關(guān)專業(yè)或?qū)ι鲜鰧I(yè)感興趣的學(xué)生； 學(xué)生需要具備線性代數(shù)、概率論、Matlab編程基礎(chǔ)

產(chǎn)出：1周在線科研+14天面授科研+5周在線論文指導(dǎo) 項(xiàng)目報(bào)告 優(yōu)秀學(xué)員獲主導(dǎo)師Reference Letter EI/CPCI/Scopus/ProQuest/Crossref/EBSCO或同等級(jí)別索引國際會(huì)議全文投遞與發(fā)表指導(dǎo)（可用于申請(qǐng)） 結(jié)業(yè)證書 成績單

項(xiàng)目背景：云計(jì)算、人工智能和比特幣所使用的去中心化分散應(yīng)用程序技術(shù)是當(dāng)下計(jì)算機(jī)科學(xué)最火熱的三個(gè)方向。并且這些技術(shù)可以被有機(jī)地結(jié)合起來，創(chuàng)建生活中的多云創(chuàng)新智能自主網(wǎng)絡(luò)（IAN）業(yè)務(wù)端和用戶端解決方案。在5G時(shí)代，各業(yè)務(wù)端的數(shù)據(jù)通量都將達(dá)到新的頂峰。為達(dá)到令用戶滿意的響應(yīng)速度、在短時(shí)間內(nèi)完成對(duì)數(shù)以萬計(jì)的數(shù)據(jù)的處理、實(shí)現(xiàn)強(qiáng)大的網(wǎng)絡(luò)服務(wù)，上述三項(xiàng)技術(shù)的結(jié)合已逐步成為必需，項(xiàng)目也將就此展開。 Cloud computing, artificial intelligence, and decentralized decentralized application technologies used by Bitcoin are three of the hottest directions in computer science today. And these technologies can be combined organically to create living multi-cloud Innovative Intelligent Autonomous Networks (IAN) business-side and user-side solutions. In the 5G era, data throughput will reach new peaks on all business ends. The combination of these three technologies is becoming necessary to achieve a satisfactory response time for users, to process tens of thousands of data in a short period of time, and to realize powerful network services, and the project will be launched accordingly.

項(xiàng)目介紹：隨著數(shù)據(jù)密集型技術(shù)出現(xiàn)，現(xiàn)代計(jì)算機(jī)系統(tǒng)對(duì)于數(shù)據(jù)存儲(chǔ)與傳輸系統(tǒng)的運(yùn)行效率與可靠性的要求也隨之增長。在本項(xiàng)目中，教授將首先介紹現(xiàn)代數(shù)據(jù)存儲(chǔ)技術(shù)的數(shù)學(xué)基礎(chǔ)，其中包括糾刪碼等信道編碼技術(shù)及其在現(xiàn)實(shí)中的應(yīng)用。而后教授將以云存儲(chǔ)系統(tǒng)為例帶領(lǐng)學(xué)生深入研究其中的算法理論與實(shí)操。學(xué)生將在項(xiàng)目結(jié)束時(shí)開發(fā)用于解決大規(guī)模AI計(jì)算的計(jì)算機(jī)算法，提交項(xiàng)目報(bào)告，進(jìn)行成果展示。With the advent of data-intensive technologies, there has been an unprecedented need to store data reliably and efficiently. In this course, we will cover the mathematical foundations of modern data storage systems, including error correction and erasure coding, and we will demonstrate how have these powerful concepts been used with great success in practical systems. We will discuss fundamental algorithms as well as modern cloud storage systems. Students will have the opportunity to learn about the latest trends in large-scale computing for AI and will grow to appreciate the universality of channel coding methods across different technologies.

項(xiàng)目大綱：算法原理與數(shù)學(xué)建模 Mathematical Modeling and Preliminaries 用于云計(jì)算的高容錯(cuò)性糾刪碼技術(shù) Erasure Coding for the Cloud 云平臺(tái)的應(yīng)用與考量 Applications and Considerations for the Modern Cloud 去中心化分布式人工智能系統(tǒng)和區(qū)塊鏈 Coded Computing for Distributed AI Systems and Blockchain 學(xué)術(shù)研討1：教授與各組學(xué)生探討并評(píng)估個(gè)性化研究課題可行性，幫助學(xué)生明晰后續(xù)科研思路 Final Project Preparation Session I 學(xué)術(shù)研討2：學(xué)生將在本周課前完成程序設(shè)計(jì)原型（prototype）及偽代碼（Pseudocode），教授將根據(jù)各組進(jìn)度進(jìn)行個(gè)性化指導(dǎo)，確保學(xué)生優(yōu)質(zhì)的終期課題產(chǎn)出 Final Project Preparation Session II 項(xiàng)目成果展示 Final Presentation