ADL99《空间大数据与人工智能》开始报名了
CCF学科前沿讲习班
The CCF Advanced Disciplines Lectures
CCFADL第99期
主题 空间大数据与人工智能
2019年6月15日-17日
北京·中国科学院计算技术研究所
随着云计算、物联网、移动计算、大数据、智慧城市等新兴信息通讯技术的迅猛发展,空间大数据及其人工智能分析已经成为一个炙手可热的重要研究方向,在诸如城市大脑、智能交通、重大科学装置数据处理、数字化战场、智能安防、工业大数据与工业互联网、应急管理及国土监测等关键应用中发挥了重要作用。
本期CCF学科前沿讲习班《空间大数据与人工智能》,讲授当前空间大数据及人工智能领域研究的最新进展和典型示范应用,旨在帮助学员快速了解和学习该领域的研究热点和前沿技术,掌握学科发展动向和重要的应用方法,开阔科研视野,增进学术交流,增强实践能力。
本期讲习班邀请到了本领域5位来自于海内外著名高校与科研机构的重量级专家学者做主题报告。他们将对空间大数据与人工智能的基础算法、关键技术、核心应用及当前热点问题进行深入浅出的讲解,并对如何开展本领域前沿技术研究等进行指导,使参加者在了解学科热点、提高理论水平的同时,掌握最新技术趋势。
特别提醒:2019年6月15日上午在本届ADL同一地点举办ACM SIGSPATIAL’2020 Pre-Workshop国际学术研讨会,欢迎本期ADL学员免费参加。
学术主任:丁治明 中国科学院软件研究所
主办单位:中国计算机学会
活动日程:
2019年6月15日下午(周六) |
|
13:30-13:45 |
开班仪式 |
13:45-14:00 |
合 影 |
14:00-17:00 |
专题讲座1:时空大数据的智能质量增强与行为预测 李 勇,清华大学,副教授 |
2019年6月16日(周日) |
|
09:00-12:00 |
专题讲座2:Data Driven Smarter Urban Transportation Systems 黄 艳,University of North Texas,教授 |
12:00-13:30 |
午 餐 |
13:30-16:30 |
专题讲座3:Privacy and Security Challenges of Spatiotemporal Data and AI 熊 莉,Emory University,教授 |
2019年6月17日(周一) |
|
09:00-12:00 |
专题讲座4:个性化推荐系统 谢 幸,微软亚洲研究院,首席研究员 |
12:00-13:30 |
午 餐 |
13:30-16:30 |
专题讲座5:面向城市安全管理的时空数据分析与智能决策 王静远,北京航空航天大学,副教授 |
16:30-17:00 |
结业式 |
l 专题讲座均包含两次15分钟课间休息 |
谢幸 微软亚洲研究院 首席研究员、中国科学技术大学 博士生导师
讲者简介:谢幸博士于2001年7月加入微软亚洲研究院,现任首席研究员,中国科技大学兼职博士生导师,以及微软-中科大联合实验室主任。他1996年毕业于中国科技大学少年班,并于2001年在中国科技大学获得博士学位,师从陈国良院士。目前,他的团队在数据挖掘、社会计算和普适计算等领域展开创新性的研究。他在国际会议和学术期刊上发表了250余篇学术论文,共被引用22000余次,H指数71,1999年获首届微软学者奖,多次在KDD、ICDM等顶级会议上获最佳论文奖,并被邀请在MDM 2019, HHME 2018, ASONAM 2017、Mobiquitous 2016、SocInfo 2015、W2GIS 2011等会议做大会主题报告。他是ACM、IEEE高级会员和CCF杰出会员,多次担任顶级国际会议程序委员会委员和领域主席等职位。他是ACM Transactions on Social Computing, ACM Transactions on Intelligent Systems and Technology、Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies (IMWUT)、Springer GeoInformatica、Elsevier Pervasive and Mobile Computing、CCF Transactions on Pervasive Computing and Interaction等杂志编委。他参与创立了ACM SIGSPATIAL中国分会,并曾担任ACM UbiComp 2011、PCC 2012、IEEE UIC 2015、以及SMP 2017等大会程序委员会共同主席。
报告题目:个性化推荐系统
报告摘要:Information overload has become a huge challenge for online users, especially for mobile users, due to the small screen size and uncomfortable inputting methods. In order to alleviate this problem, recommendation systems play an increasingly important role in Internet services, and is a constant hot topic in industry and academia. At the same time, with the rapid development of positioning, mobile and sensing technologies, large quantities of human behavioral data are now available. They reflect various aspects of human activities in the physical word, greatly improving the performance of personalized recommendation systems. In this talk, I will introduce the history of personalized recommendation systems and the challenges that are currently encountered, including the heterogeneity, sparsity, and lack of interpretability of human behavioral data. I will present how we improve the recommendation performance by leveraging the recent progress in deep learning, natural language understanding, and knowledge graph. We believe that personalized recommendation systems will continue to develop in various directions, including effectiveness, diversity, computational efficiency, and interpretability, ultimately addressing the problem of information overload.
黄 艳 University of North Texas, 教授
讲者简介:Yan Huang received her B.S. degree in Computer Science from Peking University, Beijing, China, 1997 and Ph.D. degree in, Computer Science from University of Minnesota, USA, 2003. She is currently a professor at the Department of Computer Science and Engineering and the Associate Dean for Research and Graduate Studies at College of Engineering of the University of North Texas, Denton, TX, USA. Her research interests include machine learning and data mining especially from geo-referenced datasets such as spatial intelligence, smart city, social media, and transportation data. She has been a visiting scholar of Microsoft Research Asia May – August 2011. During Fall 2011, she visited Fudan University, China. Currently, she is on the Board of Directors of The SSTD Endowment (2014-2019). She is Program Committee Chair of ACM SIGSPATIAL 2020, was the general chair of SSTD 2017, the General Chair of ACM SIGSPATIAL 2014 and 2015, and on the Executive Committee of ACM SIGSPATIAL (2010-2014). She received Distinguished Service Award from ACM SIGSpatial in 2010. Her research has been/is supported by Office of Naval Research, National Geospatial Intelligence Agency, Texas Advanced Research Program (ARP), Oak Ridge National Lab, National Science Foundation, and Texas Department of Transportation.
报告题目:Data Driven Smarter Urban Transportation Systems
报告摘要:Urban traffic gridlock is a familiar scene. With the ubiquitous availability of location enabled mobile devices and wireless communication, the time is ripe for a big data driven, dynamic, and smarter urban transportation system. In such a system, vehicles, users, and infrastructures interact with each other and are informed in real-time. They collaborate to avoidMatthew Effect; form seamless traffic flow; allow alternative and convenient means of transportation; and enable realtime ridesharing. In this talk, we will discuss algorithm and trust issues in a large scale real-time ridesharing system and bike flow prediction methods using context and new ways of location presentation learned from large scale movement data.
特邀讲者:熊 莉 Emory University, 教授
讲者简介:Li Xiong is a Professor of Computer Science and Biomedical Informatics at Emory University. She held a Winship Distinguished Research Professorship from 2015-2018. She has a Ph.D. from Georgia Institute of Technology, an MS from Johns Hopkins University, and a BS from the University of Science and Technology of China, all in Computer Science. She and her research lab, Assured Information Management and Sharing (AIMS), conduct research on algorithms and methods for big data management, data privacy and security, in the context of spatiotemporal and health data. She has published over 120 papers and received five best paper awards. She currently serves as associate editor for IEEE Transactions on Knowledge and Data Engineering (TKDE), program co-chair for ACM SIGSPATIAL 2018 and 2020, program vice-chair for IEEE International Conference on Data Engineering (ICDE) 2020, and on many program committees for data science and data security conferences. Her research is supported by National Science Foundation (NSF), AFOSR (Air Force Office of Scientific Research), National Institute of Health (NIH), and Patient-Centered Outcomes Research Institute (PCORI). She is also a recipient of Google Research Award, IBM Smarter Healthcare Faculty Innovation Award, Cisco Research Award, AT&T Research Gift, and Woodrow Wilson Career Enhancement Fellowship.
报告题目:Privacy and Security Challenges of Spatiotemporal Data and AI
报告摘要:From AI-driven medicine to self-driving cars and smart city, artificial intelligence powered by spatiotemporal data and machine learning is increasingly transforming our lives. Yet there are privacy and security pitfalls that could lead to the disclosure of sensitive data and wrong actions. These include massive collection and usage of personal data without proper privacy protection (e.g. location traces and medical records), model inversion attacks that can infer and recover sensitive training data from a trained model (e.g. reconstruct trajectories from mobility models and faces from face recognition models), to data poisoning attacks that manipulate training data at learning stage to sabotage the model (e.g. manipulate traffic reports in crowdsourcing systems to mislead traffic prediction), to adversarial example attacks that create manipulated data instances at prediction stage to deceive a model (e.g. create toxic signs to deceive self-driving cars). In this lecture, we will study these challenges and the state-of-the-art techniques towards building privacy-enhanced and robust AI in various spatiotemporal applications. We will first review the privacy challenges and introduce differential privacy as a formal privacy notion, and study its techniques and applications in various settings including centralized setting (e.g. the Census Bureau), local setting (e.g. Google and Apple), and federated setting (for multiple organizations). We will then review the adversarial attacks on machine learning algorithms including data poisoning attacks and adversarial example attacks, and study defense techniques such as adversarial example detection and adversarial training in order to build more robust AI systems.
李 勇 清华大学电子系数据科学与智能实验室 副教授
讲者简介:李勇,清华大学电子工程系副教授,博士生导师,ACM/IEEE高级会员,长期从事数据科学与智能及网络系统方面的科研工作。发表学术论文100余篇,文章引用6200余次,4次获国际会议最佳论文/提名奖,10篇论文入选ESI高被引用论文。入选国家“万人计划”青年拔尖人才、中国科协青年人才“托举工程”计划,获2016年IEEE ComSoc亚太区杰出青年学者奖,教育部科技进步一等奖、电子学会自然科学二等奖、吴文俊人工智能优秀青年奖等。
报告题目: 时空大数据的智能质量增强与行为预测
报告摘要:现代城市面临交通拥堵、能耗增加、规划落后等诸多挑战,随着智能移动终端的普及,移动终端用户在电信网络和LBSN(基于位置的社交网络)上生成了海量的时空移动数据,使得我们可以分析和研究移动用户的行为规律,为我们掌握城市基本情况、理解城市发展、提升城市运转效率提供了契机。
针对以上问题,本报告详解介绍时空移动数据的智能质量增强与用户行为建模及预测。基于大规模、多维度的时空移动数据(包括用户标识、地理位置、业务类型等信息),从移动数据处理、行为建模预测及综合平台体系构建三方面讨论包含数据接入层、数据质量增强层、数据挖掘与AI层及智慧城市应用层的时空大数据模式挖掘与行为预测关键技术及系统。
王静远 北京航空航天大学 副教授
讲者简介:北京航空航天大学计算机学院副教授,研究兴趣时空数据挖掘与智慧城市。发表学术论文30 余篇,其中包括大数据人工智能领域顶级期刊会议TKDE、KDD、ICDM、AAAI等。申请中国专利10余项,美国专利2项。承担和参与课题包括:国家自然科学基金重点项目/面上项目/青年项目、973 项目、863“智慧城市(一期/二期)”项目、国家重点研发计划等国家级科研项目多项。中国计算机学会大数据专委会委员,中国城市科学研究会大数据专委会委员,自动化学会经济管理专委会SIG委员等。
报告题目:面向城市安全管理的时空数据分析与智能决策
报告摘要:城市安全是现代城市管理的重要内容,近两年我国的一些大型城市频繁发生安全事故,给人民生命财产带来巨大威胁。时空大数据与人工智能决策为城市安全管理提供了全新的途径。本报告将从三个角度介绍城市安全管理相关的时空数据分析与智能决策技术,分别是:
1、在城市感知方面,介绍基于多源异构时空数据融合的城市感知技术;
2、在风险建模方面,介绍基于网络推断与传播分析的城市风险分析技术;
3、在管理决策方面,介绍基于可解释深度学习时空预测与决策支持技术。
报告还会以北京、天津、无锡等城市的城市安全管理业务为案例,介绍上述技术在真实城市安全管理工作中的应用。
学术主任
丁治明 中国科学院软件研究所 研究员
现任中国科学院软件研究所数据科学与数据智能研究中心主任、研究员、博导。主要研究领域为数据库与知识库系统、时空感知大数据系统、物联网与智慧城市等。曾工作于德国时空数据管理领域的国际知名专家Ralf Hartmut Güting教授团队;长期在中国科学院系统(计算所、软件所)学习与工作;2014年8月通过海内外院长招聘担任北京工业大学计算机学院院长。2018年4月重新引进回中国科学院软件研究所工作至今。2016年获国务院政府特殊津贴、青海省“千人计划”领军人才,2018年获北京市特聘教授。是中国计算机学会(CCF)数据库专委会委员、CCF大数据专委会委员、中国健康大数据产业联盟常务理事,担任IEEE Intelligent Transportation System Society的社会交通专委会主席、ACM SIGSpatial China Chapter的副主席,担任国际刊物IEEE Transactions on Intelligent Transportation Systems、IEEE Intelligent Transportation Systems Magazine的编委。在国内外学术刊物发表论文130余篇,出版3部学术专著,获6项发明专利、8项软件著作权,制定国家标准1项,曾获得北京市科技进步奖、国家优秀科技信息成果奖等荣誉。
时间:2019年6月15-17日
地点:北京(中国科学院计算技术研究所,北京市海淀区科学院南路6号)
注册费:CCF会员3000,非会员3600元,食宿交通自理。
西部五所高校(新疆大学、青海大学、贵州大学、云南大学、宁夏大学)每个学校有两个免报名费名额。
报名方式:点击以下链接或扫描(识别)二维码报名。报名链接:https://conf2.ccf.org.cn/adl99
缴费方式:
在报名系统中在线缴费或者通过银行转账
银行转账(支持网银、支付宝):
开户行:北京银行北京大学支行
户名:中国计算机学会
账号:0109 0519 5001 2010 9702 028
请务必注明:姓名+ADL99
报名缴费后,报名系统中显示缴费完成,即为报名成功。
联系人:李红梅 邮箱: adl@ccf.org.cn 电话:18810669757
本期ADL招募志愿者:
1、班长和副班长:招募班长、副班长各一名。
要求:志愿为学员服务;配合讲师,带领学员共同完成学习任务。
2、通讯员:招募一名通讯员,需要在课程结束时完成一篇新闻稿,内容围绕本期课程内容、讲师与学员互动情况等。要求文笔流畅,图文并茂。稿件将在CCF官网、官微和中国计算机学会通讯上署名发表(有稿费)。
3、摄影志愿者:所有学员均可提交与本期ADL有关照片,课程结束后选出精彩照片10张,这10张照片的摄影者为本期ADL摄影志愿者。
4、课余“主题活动”招募人:若干名。负责招募有共同爱好的学员,并组织活动,丰富学员课余生活。
5、为增进学员和讲者之间的交流,创造更多的机会,CCF将从本期ADL开始增加互动环节,组织晚餐会活动。将从前50名报名的学员中抽取10人与讲者共进晚餐(自愿,费用AA)。
为感谢志愿者的辛勤付出,CCF为每位志愿者准备精美礼品一份。志愿者今后再次参加ADL,可以享受报名注册费优惠。