報(bào)告地點(diǎn)：西北工業(yè)大學(xué)長(zhǎng)安校區(qū)計(jì)算機(jī)學(xué)院105會(huì)議室
報(bào)告時(shí)間：2019年5月6日上午10：30-11：30
報(bào)告人：香港中文大學(xué)John C.S. Lui教授, IEEE/ACM Fellow
邀請(qǐng)人：褚偉波副教授
報(bào)告題目：一類(lèi)基于在線(xiàn)學(xué)習(xí)的網(wǎng)絡(luò)多路徑選擇方法
報(bào)告人簡(jiǎn)介:
呂自成教授目前是香港中文大學(xué)計(jì)算機(jī)科學(xué)與工程系的李卓敏榮譽(yù)教授。他的研究方向聚焦于機(jī)器學(xué)習(xí)在網(wǎng)絡(luò)科學(xué)、網(wǎng)絡(luò)經(jīng)濟(jì)學(xué)、網(wǎng)絡(luò)/系統(tǒng)安全、大規(guī)模分布式系統(tǒng)和性能測(cè)評(píng)理論方面的研究與應(yīng)用。呂教授獲得了諸多教學(xué)與科研方面的獎(jiǎng)項(xiàng)，包括香港中文大學(xué)校長(zhǎng)模范教學(xué)獎(jiǎng)和香港中文大學(xué)職員杰出研究獎(jiǎng)（2011-2012）。他獲得了IFIP WG 7.3 Performance 2005, IEEE/IFIP NOMS 2006，SIMPLEX'14，ACM RecSys’17等重要國(guó)際會(huì)議的最佳學(xué)術(shù)論文獎(jiǎng)以及ACM Mobihoc’18和ASONAM’17會(huì)議的最佳論文提名獎(jiǎng)。他是IFIP WG 7.3，ACM，IEEE等多個(gè)重要協(xié)會(huì)的會(huì)士，Croucher基金會(huì)的高級(jí)研究專(zhuān)家，以及現(xiàn)任的ACM SIGMETRICS會(huì)議主席。呂教授的個(gè)人興趣包括電影和閱讀。
報(bào)告摘要:
過(guò)去十年間，接入計(jì)算機(jī)網(wǎng)絡(luò)的主機(jī)數(shù)量呈現(xiàn)出爆炸式增長(zhǎng)。多種用于主機(jī)間數(shù)據(jù)傳輸?shù)亩嗦窂絽f(xié)議被相繼提出。然而，當(dāng)前的數(shù)據(jù)傳輸多路徑協(xié)議由于忽略了網(wǎng)絡(luò)傳輸延遲、可用帶寬和數(shù)據(jù)丟包等隨機(jī)特性而使它們?cè)趹?yīng)用上收到了極大的限制。另外，許多應(yīng)用對(duì)網(wǎng)絡(luò)傳輸延遲、鏈路帶寬和丟包率等有一定要求。本報(bào)告將介紹一種基于網(wǎng)絡(luò)特性在線(xiàn)學(xué)習(xí)的多路徑選取框架，用于滿(mǎn)足不同應(yīng)用對(duì)數(shù)據(jù)傳輸?shù)男枨?。具體地，將介紹一類(lèi)用于分別滿(mǎn)足最大傳輸延遲、網(wǎng)絡(luò)帶寬和丟包率約束的在線(xiàn)學(xué)習(xí)多路徑選取算法，并在理論上確保算法具有次線(xiàn)性的regret和violation兩個(gè)關(guān)鍵性能指標(biāo)。
Biography:
John C.S. Lui is currently the Choh-Ming Li Professor of the Computer Science & Engineering Department at The Chinese University of Hong Kong. His current research interests are in machine learning on network sciences, network economics, network/system security (e.g., cloud security, mobile security, ...etc), large scale distributed systems and performance evaluation theory. John received various departmental teaching awards and the CUHK Vice-Chancellor's Exemplary Teaching Award, as well as the CUHK Faculty of Engineering Research Excellence Award (2011-2012). He is a co-recipient of the IFIP WG 7.3 Performance 2005, IEEE/IFIP NOMS 2006 and SIMPLEX'14 Best Paper Awards, ACM RecSys’17 best paper award and best paper runner-up in ACM Mobihoc’18 and ASONAM’17. He is an elected member of the IFIP WG 7.3, Fellow of ACM, Fellow of IEEE, Senior Research Fellow of the Croucher Foundation and is currently the chair of the ACM SIGMETRICS. His personal interests include films and general reading.
Title:
An Online Learning Multi-path Selection Framework for Multi-path Transmission Protocols
Abstract:
In the last decade, we have witnessed a tremendous growth of inter-connectivity among hosts in networks. Many new data transmission protocols have been developed to enable multi-path data transmissions between two hosts. However, the existing multi-path transmission protocol designs are limited as they neglect the stochastic nature of the metrics of the paths, e.g., latency, available bandwidth, and packet loss. Moreover, there are different design requirements in the applications, such as low latency, bandwidth throttling, and low loss rate in data delivery. In this talk, we propose a flexible online learning multi-path selection (OLMPS) framework to select multiple paths by learning the stochastic metrics of the paths and meeting the design requirements of the applications. Specifically, we design a set of novel online learning algorithms in the OLMPS framework for three different applications -- maxRTT constrained, bandwidth constrained, and loss rate constrained, multi-path selection, to select paths and satisfy the requirements. We prove that the algorithms can provide theoretical guarantees on both sublinear regret and sublinear violation in our OLMPS framework.
