讲座名称:基于机器学习的特定应用NoC的有效设计空间探索
讲座时间:2019-06-22 10:00:00
讲座地点:西电北校区东大楼221学术报告厅
讲座人:Ulf Schlichtmann
讲座人介绍:
Ulf Schlichtmann,博士,教授,德国慕尼黑工业大学全职讲席教授,慕尼黑工业大学EDA实验中心主任,曾/现担任德国英飞凌科技部门主管以及最高管理层成员,德国国家科学与工程院成员,新加坡慕尼黑工业大学研究中心主任。2007年至2013年担任慕尼黑工业大学电子与计算机工程学院副院长和院长,2016年担任慕尼黑工业大学董事会成员和教授任命委员会委员。Schlichtmann教授发表了250多篇顶级期刊和会议论文,获得了多项会议的最佳论文提名和最佳论文奖,包括Design, Automation and Test in Europe (DATE) 2017最佳论文奖。
讲座内容:
In this talk, I will give a brief overview of the ongoing research topics in the Chair of Electronic Design Automation (EDA) at Technical University of Munich (TUM), covering analysis and optimization of electronic circuits and systems as well as emerging technologies such as biochips and photonic interconnect. Thereafter, our research on Multi-Processor Systems-on-Chip (MPSoCs) with an increasing number of heterogeneous processing elements will be presented in detail. Since the traffic requirements on the Network on Chip (NoC) become more and more unbalanced, the NoC network architecture must be customized according to the needs of the application at hand. In this talk, I will introduce two approaches for the acceleration of design space exploration (DSE) through machine learning techniques.
主办单位:微电子学院