学 术

分享到微信 ×
打开微信“扫一扫”
即可将网页分享至朋友圈
学者论坛:Model Based, Explainable Deep Learning Via Algorithm Unrolling
文:教师发展中心 来源:党委教师工作部、人力资源部(教师发展中心) 时间:2024-07-08 615

教师发展中心“学者论坛”活动特别邀请美国宾夕法尼亚州立大学Vishal Monga教授来校作学术交流。具体安排如下,欢迎广大师生参加。

一、主 题:Model Based, Explainable Deep Learning Via Algorithm Unrolling

二、时 间:2024年7月8日(周一) 14:00-16:00

三、地 点:清水河校区宾诺咖啡

四、主讲专家:美国宾夕法尼亚州立大学Vishal Monga教授

五、主持人:信息与通信工程学院 余显祥 副教授

六、内容简介:

The talk will introduce algorithm unrolling, a rigorous and methodical framework for designing efficient, explainable neural networks by leveraging iterative optimization approaches that have long provided the analytical foundations of signal processing and computational imaging. Applications and associated analytical results will be demonstrated in blind and non-blind image deconvolution.

七、主讲人简介:

Professor Vishal Monga has been with the Department of Electrical Engineering at Pennsylvania State University since 2009. He is a member of the IEEE Technical Committees on SAM, Computational Imaging, and Bioimaging and Signal Processing. Dr. Monga was inducted into the National Academy of Inventors in 2022 and holds 45 U.S. patents. He earned his Ph.D. from the University of Texas at Austin, receiving a National Science Foundation CAREER Award. From 2005 to 2009, he worked as an imaging scientist at Xerox Research Labs and has been a visiting researcher at Microsoft Research and the University of Rochester. He received the 2019 PSEAS Distinguished Research Award and the 2022 PSEAS Excellence in Research Award. From 2017-19, he served as the Technical Direction Chair of the IEEE Image, Video, and Multidimensional Signal Processing Technical Committee. He has served on the editorial boards of several journals, including IEEE TIP, JSTSP, TCSVT, and SPL etc, and is a founding editorial board member of Frontiers in Signal Processing. 

八、主办单位:教师发展中心

  承办单位:信息与通信工程学院   生命科学与技术学院


编辑:刘瑶  / 审核:李果  / 发布:陈伟