学术报告201622期

报告题目：Integrated  Learning and Control for Machine Intelligence

报告时间：6月27日下午15:30至16:30

报告地点：南一楼中311

报  告  人：何海波教授

邀  请  方：“多谱信息处理技术”国家级重点实验室

Abstract:

With  the recent development of brain research and modern technologies, scientists and  engineers will hopefully find efficient ways to develop brain-like intelligent  systems that are highly robust, adaptive, scalable, and fault tolerant to  uncertain and unstructured environments. Yet, developing such truly intelligent  systems requires significant research on both fundamental understanding of brain  intelligence as well as complex engineering design. This talk aims to present  the recent research developments in computational intelligence to advance the  machine intelligence research and explore their wide applications in complex  cyber physical systems across different domains.

Specifically,  this talk will focus on a new adaptive dynamic programing (ADP) framework with  rich internal goal representation for improved learning and optimization  capability over time. This architecture integrates an internal goal generator  network to provide a more informative and detailed internal value representation  to support the decision-making process. Compared to the existing ADP approaches  with a manual or “hand-crafted” reinforcement signal design, our approach can  automatically and adaptively develop the internal reinforcement signal over  time, therefore improving the learning and control performance. This internal  goal network can also be designed in a hierarchical way, to provide a  multi-level value representation. Under this ADP framework, I will present  numerous applications including smart grid and human-robot interaction, to  demonstrate its broader and far-reaching applications. As a multi-disciplinary  research topic, I will also briefly discuss the future research challenges and  opportunities in this field.

Brief  Bio:

Haibo  He is the Robert Haas Endowed Chair Professor and the Director of the  Computational Intelligence and Self-Adaptive (CISA) Laboratory at the University  of Rhode Island, Kingston, RI, USA. His primary research interests include  computational intelligence and its applications to complex systems. He has  published one sole-author book (Wiley), edited 1 book (Wiley-IEEE) and 6  conference proceedings (Springer), and authored/co-authors over 200  peer-reviewed journal and conference papers, including several highly cited  papers in IEEE Transactions on Neural Networks and IEEE Transactions on  Knowledge and Data Engineering, Cover Page Highlighted paper in IEEE  Transactions on Information Forensics and Security, and Best Readings of the  IEEE Communications Society. He has delivered more than 40 invited talks around  the globe, including numerous keynote and plenary talks. His research has been  covered by numerous national and international medias, such as IEEE Smart Grid  Newsletter, The Wall Street Journal, China Central Television (CCTV), Providence  Journal, Providence Business News, among others.

He  has served the IEEE Computational Intelligence Society (CIS) at various  capacities, including Chair of IEEE CIS Emergent Technologies Technical  Committee (ETTC) (2015), Chair of IEEE CIS Neural Networks Technical Committee  (NNTC) (2013 and 2014), Vice Chair of IEEE CIS Adaptive Dynamic Programming and  Reinforcement Learning (ADPRL) Technical Committee (2012 and 2013), IEEE-USA  Energy Policy Committee, among others. He was the General Chair of 2014 IEEE  Symposium Series on Computational Intelligence (IEEE SSCI’14), Technical Program  Co-Chair of 2015 International Joint Conference on Neural Networks (IJCNN’15),  Program Co-Chair of 2014 International Joint Conference on Neural Networks  (IJCNN’14), among others. He serves as an Associate Editor of IEEE Transactions  on Neural Networks and Learning Systems (2010 to 2015), IEEE Computational  Intelligence Magazine (2015), IEEE Transactions on Smart Grid (2010 to present),  and IEEE Internet of Things Journal (2016 to present). Currently, He is the  Editor-in-Chief of IEEE Transactions on Neural Networks and Learning  Systems.

He  was a recipient of the IEEE International Conference on Communications (ICC)  “Best Paper Award” (2014), IEEE CIS “Outstanding Early Career Award” (2014),  National Science Foundation “Faculty Early Career Development (CAREER) Award”  (2011), Providence Business News (PBN) “Rising Star Innovator” Award (2011), and  “Best Master Thesis Award” of Hubei Province, China  (2002).