Hello everyone,
I'm Mingxing Zhang, currently a Ph.D. student at Tsinghua University in
China (majoered in Computer Science in Institute for High Performance
Computing).
If possible, I want to apply for the idea "Coordinated descent in
linear models beyond squared loss (eg Logistic)" in GSoC. This project is
very attractive for me, and I think I'm suitable to fulfill it.
First, I'm familiar with logistic regression and interesting in its
optimization. Actually, during the last 9 months, I interned in System
Research Group <http://research.microsoft.com/en-us/groups/sr/> of
Microsoft Research Asia on a similar problem. Under Chuntao
Hong<http://research.microsoft.com/en-us/people/chhong/>'s
mentoring, I re-write Microsoft Search Technology Center (aka bing.com)'s
L-BFGS implementation with MPI and finally reduce its execution time from 3
days per 40 iteration to less than 3 hours with even less machines (2TB
data in total, I use 20 mechines, eache of them has 16 cores).
Second, I love python! I use python in almost all the projects if speed
is not that critical, it saves me a lot of time. And I also cooperated with
friends and created http://synclee.net/ with Django.
Third, I'm good at algorithms. I have attended to ACM International
Collegiate Programming Contest and ranked 36 at 2012 World Finals in Poland
(a list of my awards can be find
here<http://www.linkedin.com/profile/view?id=80873276>).
Finally, sorry for applying in such a later time. I accidentaly comed
accross the news of GSoC in weibo.com (twitter of China) today, and taken
this a good oppotunity to get involve in open source projects, which help
me a lot. After skimming the idea lists of several projects, I think that
this is the most suitable idea for me.
So I wonder that wether this position is still available, or is there
anything I can do to maximize my oppotunity?
Thanks a lot.
--
章明星
电话：010-62797143
清华大学计算机科学与技术系
高性能计算研究所 FIT楼3-122
北京，100084
Mingxing Zhang
Tel.: +86-10-62797143
Room 3-122, FIT Building
Institute for High Performance Computing,
Department of Computer Science and Technology,
Tsinghua University, Beijing 100084, China
------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_apr
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
