Difference between revisions of "User:Xiaoqiy"
| (3 intermediate revisions by the same user not shown) | |||
| Line 4: | Line 4: | ||
[[File:XiaoqiYin.png]] | [[File:XiaoqiYin.png]] | ||
| − | Hi! My name is | + | Hi! My name is Xiaoqi Yin, or 阴小骐 in Chinese. You can call me Philip! |
== Who I am and why I'm here == | == Who I am and why I'm here == | ||
| Line 10: | Line 10: | ||
I graduated from Department of Automation, Tsinghua University in 2010. Currently I am a second-year PhD student in ECE, CMU. My advisor is Prof. Bruno Sinopoli (http://www.ece.cmu.edu/~brunos/). | I graduated from Department of Automation, Tsinghua University in 2010. Currently I am a second-year PhD student in ECE, CMU. My advisor is Prof. Bruno Sinopoli (http://www.ece.cmu.edu/~brunos/). | ||
| − | I am interested in machine learning in general. | + | I am interested in machine learning in general. I think logic structure, which is the nature of the knowledge of our world, can help machine better learn about the world. However, few of the main-stream machine learning techniques focus on how to learn and apply relationships based on large-amount of data. Although HMM has been well developed as one of the most popular techniques to analyze sequential data, the logic structure of it has been fixed before learning. I think new relational models and learning methods should be developed to understand the highly networked data which comes from the Internet. |
| − | + | My past research include collective classification on social networks and non-intrusive load monitoring, both of which are related to "structured prediction" methods! | |
| − | |||
| − | |||
| − | |||
Latest revision as of 23:40, 5 September 2011
Philip(Xiaoqi) Yin
Hi! My name is Xiaoqi Yin, or 阴小骐 in Chinese. You can call me Philip!
Who I am and why I'm here
I graduated from Department of Automation, Tsinghua University in 2010. Currently I am a second-year PhD student in ECE, CMU. My advisor is Prof. Bruno Sinopoli (http://www.ece.cmu.edu/~brunos/).
I am interested in machine learning in general. I think logic structure, which is the nature of the knowledge of our world, can help machine better learn about the world. However, few of the main-stream machine learning techniques focus on how to learn and apply relationships based on large-amount of data. Although HMM has been well developed as one of the most popular techniques to analyze sequential data, the logic structure of it has been fixed before learning. I think new relational models and learning methods should be developed to understand the highly networked data which comes from the Internet.
My past research include collective classification on social networks and non-intrusive load monitoring, both of which are related to "structured prediction" methods!
