[mlpack] GSoC 2016 - Fast k-centers Algorithm & Implementation

[Borja Menéndez Moreno]

Hello,

I am Borja Menéndez, a PhD student in Computer Science starting his third
year of studies from Madrid, Spain. In particular, my field of study is
optimization problems. In my thesis I'm working on combinatorial problems
related to retrieving orders from warehouses. You can see my first JCR
paper in [1] for more info about this.

As you can imagine, I'm really interested in the project titled "Fast
k-centers Algorithm & Implementation". I find this optimization problem
truly challenging and funny to work with. In the last few weeks I'm being
attracted to Machine Learning and I think this project is a good start for
me since it is a good mixture between optimization problems and machine
learning. During my master I worked with an optimization problem relatively
similar to this one: MaxMin Diversity Problem. I know it is not the same
problem but, probably, the ideas I used for that problem could work here.

Currently, and from the last two years, I'm working with Java, but I also
have programmed my whole master thesis in C++ (more than 12.000 lines of
code) in an open source project called JdeRobot, a robotics framework. You
can see my contribution to the project in [2] (my github username is
bmenendez [3]).

Are you open to try other algorithms (different from the Dual-Tree proposed
in the ideas page) such as Variable Neighborhood Search / GRASP / Tabu
Search to address the problem? Are you planning to write a paper if the
obtained solutions are good enough to do it? Do you have a set of problem
instances to start working on this problem and trying some naïve ideas?

I also wanted to congratulate for your work in mlpack since it is an
extremely easy to install library and your documentation is remarkably
complete.

Best regards,

Borja.

[1] - http://www.sciencedirect.com/science/article/pii/S0305054816300168
[2] - https://github.com/RoboticsURJC/JdeRobot/graphs/contributors
[3] - https://github.com/bmenendez
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cc.gatech.edu/pipermail/mlpack/attachments/20160313/d9c3895a/attachment-0002.html>