首页 | 综合通知 | 教务教学 | 学术交流 | 留学信息 | 科研管理 | 文献服务 | 招标信息 | 网络服务 
当前位置: 校园公告>>学术交流>>正文
【学术会议】“人机智能”国际前沿交叉学科论坛会议通知
2018-04-16 15:02   审核人:   (点击: )

1、论坛简介

人机物融合正成为新一代计算技术的重要特征与主要趋势,人类和机器智能的融合是关键。人机智能融合不是人类和机器智能在物理上的简单混合,而是通过融合产生化学反应,迸发出更多更强的智慧,通过不同粒度智能的深度交融、共同演进,实现人类和机器智能的共融共生。人机共融智能的关键特性包括个体智能融合、群体智能融合、智能共同演进,目标是人类智能和机器智能互相适应,彼此支持,相互促进,实现智能的共同演进和优化。

近年来,在国家“973”计划、国家杰出青年科学基金、自然基金重点项目等课题支持下,计算机学院围绕人机物融合计算、协同感知、群智计算等人机智能前沿方向开展了系列研究工作,取得了若干研究成果。为了进一步推动相关研究,拟于2018年4月17-18日召开“人机智能”前沿交叉学科论坛,邀请多位IEEE/IET会士,围绕“人机智能”的研究挑战、基础理论与方法等开展深入研讨与交流。

2、论坛时间与地点

论坛时间:2018年4月17-18日

论坛地点:西北工业大学友谊校区国际会议中心第二会议室

3、论坛议程

时间

(4月18日)

会议内容

地点

主持人

8:30-9:00

开幕式(领导致辞、合影)

国际会议中心

第二会议室

於志文

9:00-9:30

加拿大西蒙弗雷泽大学Ljiljana Trajkovic教授

报告题目:Machine Learning for Complex Networks

9:30-10:00

台湾国立中兴大学蔡清池(Ching-Chih Tsai)教授

报告题目:Intelligent Adaptive Learning Control Methods with Applications to Mobile Robots and Multirobots

10:00-10:30

茶歇

10:30-11:00

南方科技大学Hisao Ishibuchi教授

报告题目:Fair Performance Comparison of Evolutionary Multi-Objective and Many-Objective Optimization Algorithms

国际会议中心

第二会议室

史豪斌

11:00-11:30

台湾国立科技大学苏顺丰(Shun-Feng Su)教授

报告题目:Decomposed Fuzzy Systems

11:30-12:00

台湾中山大学黄国胜(Kao-Shing Hwang)教授

报告题目:以强化学习实现适应性视觉伺服

4、论坛组委会

论坛主席

    於志文教授    计算机学院党委书记、国家杰青、“万人计划”领军人才

论坛程序委员会主席

    史豪斌副教授  计算机学院信息安全与电子商务技术系 系主任

论坛执行委员会主席

    潘炜副教授    计算机学院

    王柱副教授    计算机学院

5、论坛专家介绍

Ljiljana Trajkovic

School of Engineering Science

Simon Fraser University, Canada

IEEE Follow/Senior Past President (2018–2019) of the IEEE Systems, Man, and Cybernetics Society

Talk Title: Machine Learning for Complex Networks

Abstract:

Collection and analysis of data from deployed networks is essential for understanding modern networks. Traffic traces collected from various deployed communication networks and the Internet have been used to characterize and model network traffic, analyze Internet topologies, and classify network anomalies. Data mining and statistical analysis of network data are often employed to determine traffic loads, analyze patterns of users' behavior, and predict future network traffic. Spectral graph theory has been applied to analyze various topologies of complex networks and capture historical trends in their development. Recent machine learning techniques have proved valuable for predicting anomalous traffic behavior and for classifying anomalies in complex networks. Further applications of these tools will help improve our understanding of the underlying mechanisms that govern the behavior of complex networks such as the Internet, social networks (Facebook, LinkedIn, Twitter, Internet blogs, forums, and websites), power grids, gene regulatory networks, neuronal systems, food webs, social systems, and networks emanating from augmented and virtual reality platforms. They will also help improve performance of these networks and enhance their security.

Biography:

Ljiljana Trajkovic is currently a Professor in the School of Engineering Science at Simon Fraser University, Burnaby, British Columbia, Canada. She received the Dipl. Ing. degree from University of Pristina, Yugoslavia, in 1974, the M.Sc. degrees in electrical engineering and computer engineering from Syracuse University, Syracuse, NY, in 1979 and 1981, respectively, and the Ph.D. degree in electrical engineering from University of California at Los Angeles, in 1986.

Dr. Trajkovic serves as IEEE Division Delegate-Elect/Director-Elect (2018). She serves as Senior Past President (2018–2019) of the IEEE Systems, Man, and Cybernetics Society and served as Junior Past President (2016–2017), President (2014–2015), President-Elect (2013), Vice President Publications (2012–2013 and 2010–2011), Vice President Long-Range Planning and Finance (2008–2009), and a Member at Large of its Board of Governors (2004–2006). She is a Professional Member of IEEE-HKN and a Fellow of the IEEE.

老師照片

蔡清池(Ching-Chih Tsai)

Department of Electrical Engineering

National Chung-Hsing University

IEEE Fellow/IET Fellow /CACS Fellow

Talk Title: Intelligent Adaptive Learning Control Methods with Applications to Mobile Robots and Multirobots

Abstract:

Intelligent adaptive learning control (IALC) has been widely investigated and applied for frontier mobile robotics. By incorporating the merits of predictive control, fuzzy modeling and recurrent fuzzy neural networks, this talk will present you three novel ILC framework or paradigms for a class of mobile robots and multirobots in order to achieve desired motion control and navigation. In the outset of the talk, some literature reviews are first mentioned, and then three novel ILC methods and their applications to wheeled robots and multirobots are briefly highlighted and experimentally demonstrated. First, some stable IALC paradigm using feedforward and recurrent fuzzy neural-network structures are discussed which have been well employed to motion control of uncertain wheeled mobile platforms. Second, some new consensus formation control paradigm using IALC are proposed for a class of mobile multirobots. Third, one IALC paradigm using deep learning is simply presented and shown effective for intelligent mobile robots and multirobots. Last but not least, some perspective topics are recommended for future research.

Biography:

Ching-Chih Tsai is currently a Distinguished Professor in the Department of Electrical Engineering, National Chung-Hsing University, Taichung, Taiwan, where he served the Chairman in the Department of Electrical Engineering from 2012 to 2014. He received the Diplomat in Electrical Engineering from National Taipei Institute of Technology, Taipei, Taiwan, ROC, the MS degree in Control Engineering from National Chiao Tung University, Hsinchu, Taiwan, ROC and the Ph.D degree in Electrical Engineering from Northwestern University, Evanston, IL, USA, in 1981, 1986 and 1991, respectively.

Dr. Tsai served as the Chair, Taipei Chapter, IEEE Control Systems Society, from 2000 to 2003, and the Chair, Taipei Chapter, IEEE Robotics and Automation Society from 2005 to 2006. He has served as the Chair, Taichung Chapter, IEEE Systems, Man, and Cybernetics Society since 2009, the Chair of IEEE SMC Technical Committee on intelligent learning in control systems since 2009, the President of Robotics Society of Taiwan since 2016, the steering committee of Asian Control Association since 2014, a BOG member of IEEE Nanotechnology council since 2012, the Vice President of International Fuzzy Systems Association since 2015, and a BOG member of the IEEE SMCS since 2017. He is an IEEE Fellow, an IET Fellow and a CACS Fellow.

Hisao Ishibuchi

Chair Professor

Department of Computer Science and Engineering

Southern University of Science Technology

IEEE Fellow

Talk Title: Fair Performance Comparison of Evolutionary Multi-Objective and Many-Objective Optimization Algorithms

Abstract:

Evolutionary multi-objective optimization (EMO) has been an active research area in the field of evolutionary computation. Various EMO algorithms have been proposed in the literature. Their main characteristic feature in comparison with other optimization techniques is that a set of non-dominated solutions (instead of a single optimal solution) is obtained by their single run. This means that the comparison of different EMO algorithms needs an evaluation mechanism of non-dominated solution sets. The focus of this talk is how to compare different EMO algorithms using performance indicators of non-dominated solution sets. In this talk, we first briefly explain evolutionary multi-objective and many-objective optimization algorithms. Next we explain the hypervolume (HV) and the inverted generational distance (IGD), which are the most frequently-used performance indicators. Then we show difficulties of each performance indicator. For example, it is explained that HV-based performance comparison results of different EMO algorithms depends on the specification of a reference point for HV calculation. It is also explained that a set of uniformly distributed non-dominated solutions over the entire Pareto front is not the best distribution of solutions for IGD maximization. Finally, we discuss the parameter specifications for fair performance comparison of different EMO algorithms.

Biography:

Hisao Ishibuchi is currently with Department of Computer Science and Engineering, SUSTech, Shenzhen, China as a Chair Professor. He received the BS and MS degrees from Kyoto University in 1985 and 1987, respectively. In 1992, he received the Ph. D. degree from Osaka Prefecture University where he has been a professor since 1999.

Dr. Ishibuchi was the IEEE CIS Vice-President for Technical Activities (2010-2013) and an IEEE CIS Distinguished Lecturer (2015-2017). Currently, he is the President of the Japan EC Society (2016-2018), the Editor-in-Chief of IEEE CI Magazine (2014-2019) and Journal of Japan EC Society (2014-2018), an IEEE CIS AdCom member (2014-2019). He is also an Associate Editor of IEEE TEVC (2007-2018), IEEE Access (2013-2018) and IEEE TCyb (2013-2018). He is an IEEE Fellow. In 2018, he was selected in the “Recruitment Program of Global Experts for Foreign Experts” known as the “Thousand Talents Program.”

苏顺丰(Shun-Feng Su)

Department of Electrical Engineering

National Taiwan University of Science and Technology

IEEE Fellow/CACS fellow

Talk Title: Decomposed Fuzzy Systems

Abstract:

In the talk, a novel fuzzy structure termed as the decomposed fuzzy system (DFS) is proposed to act as the fuzzy approximator. The proposed structure is to decompose each fuzzy variable into layers of fuzzy systems and each layer is to characterize one traditional fuzzy set. Similar to forming fuzzy rules in traditional fuzzy systems, layers from different variables will form the so-called component fuzzy systems. The structure of DFS is proposed to facilitate minimum distribution learning effects among component fuzzy systems so that the learning can be very efficient. It can be seen from our experiments that even when the rule number increases, the learning time in terms of cycles is still almost constant. It can also be found that the function approximation capability and learning efficiency of the DFS are much better than that of the traditional fuzzy systems when employed in adaptive fuzzy control systems. Besides, in order to further reduce the computational burden, a simplified DFS is proposed in this study to satisfy possible real time constraints required in many applications. From our simulation results, it can be seen that the simplified DFS can perform fairly with a more concise decomposition structure. Furthermore, when used in modeling, the proposed DFS not only can have much faster convergent speed, but also can achieve a smaller testing error than those of other fuzzy systems.

Biography:

Shun-Feng Su is now a Chair Professor of the Department of Electrical Engineering, National Taiwan University of Science and Technology, Taiwan, R.O.C. He received the B.S. degree in electrical engineering, in 1983, from National Taiwan University, Taiwan, R.O.C., and the M.S. and Ph.D. degrees in electrical engineering, in 1989 and 1991, respectively, from Purdue University, West Lafayette, IN.

Dr. Su is now the past president of the International Fuzzy Systems Association. He also serves as a board member of various academic societies. He acted as General Chair, Program Chair, or various positions for many international and domestic conferences. Dr. Su currently serves as Associate editors of IEEE Transactions on Cybernetics, IEEE Transactions on Fuzzy Systems, and IEEE Access, a subject editor (Electrical Engineering) of the Journal of the Chinese Institute of Engineers, and the Editor-in-Chief of International Journal of Fuzzy Systems. He is an IEEE Fellow and CACS fellow.

黄国胜(Kao-Shing Hwang)

Department of Electrical Engineering

National Sun Yat-sen University

IET Fellow

Talk Title:以强化学习实现适应性视觉伺服

Abstract:

强化学习算法中奖惩函数的设计通常都是目标为导向的透过代理人以尝试错误的方法来学习完成目标。演讲主题主要报告在影像视觉伺服系统里引入加强式学习中的Q学习法设计一个智能型增益控制器,并应用于机器手臂控制。利用图像处理算法进行目标影像与当前影像的特征撷取后,计算影像特征向量间的误差距离,此误差距离将泛化后形成Q学习的状态空间。而动作空间将由控制增益量组成,根据手臂从影像取得的状态,利用贪婪算法选择适当的动作,进行手臂的移动控制。为进一步增加手臂逼近目标影像精确度,本报告基于原本的动作空间引入一衰减值,当目前特征误差小于一定量时,每一动作将附上一衰减值减少控制量,使得手臂于目标位置附近时能更加精确与稳定。透过本报告提出的方法使机器手臂在进行影像视觉伺服的过程中,可以解决固定控制增益值过大时造成控制系统过冲,以及控制增益值过小时使手臂移动速度过于缓慢的问题。由于Q学习法不需要事先拥有任何环境相关的知识,即可进行学习,适合应用在决策控制系统的问题上。Q学习法藉由学习代理人与环境互动取得报酬,互动的同时Q学习会根据报酬值的强弱调整策略,当经过多次且长时间与环境互动,累积一定数量的经验,最后代理人会学习到一组最佳策略。经学习后的控制增益值能使系统稳定的达到目标状态位置,且有效的减少达到目标状态的单位时间。

Biography:

黄国胜教授现任职于台湾中山大学电机系。其系于1993年获美国西北大学计算机工程博士学位,旋及被台湾中正大学电机系延揽回国任教并继续从事于机器人科学技术方面的研究。这十几年担任过中正大学电算中心组长、代主任、电机系系主任、光机电整合研究所所长、国科会自动控制学门规划及复审委员、自动化学门复审委员。由于其学术表现受认同,因此成为中华民国自动控制学会会士、欧洲电子电机学会(IET)院士、IEEE Trans. on Cybernetics、IEEE/ACM Trans. on Mechatronics编辑、IEEE/CAA Journal of Automatica Sinica编辑,以及International Journal of Fuzzy Systems等编辑也曾受邀为中国上海交通大学荣誉客座教授。黄国胜教授的研究领域与兴趣包含了机器人路径规划、机器足球员系统、强化合作学习系、以及群组机器人任务合作。

关闭窗口
最新公告
西工大伦敦玛丽女王大学工程学院2017...
感应耦合等离子刻蚀系统(ICP)免费试...
关于征集西北工业大学80周年校庆摄影...
多功能物性测量系统(PPMS)免费试运...
南山艺堂---“古方香囊”创意体验活动...
西北工业大学校史资料征集启事
关于开展全校形象标识检查与规范的通知
分析测试中心友谊校区分中心首套设备...
校分析测试中心原子力显微镜(AFM)开...
中国政府奖学金高校研究生项目(国际...
西工大主页  招聘信息  就业信息  本科招生  研招信息
版权所有©西北工业大学  地址:西安市友谊西路127号 邮编:710072