[mlpack] About the GSoC '17 "Reinforcement learning" project

[Marcus Edel]

Hello Vladimir,

> first of all I would like to congratulate to mlpack for being accepted as a
> participating organization in the Google Summer of Code 2017.


Thanks and welcome!

> However, I never had a chance to implement a solution for some of the concepts
> that requires analytical thinking and to write solid amount of code. Therefore,
> I would like to take part in the development of mlpack during GSoC, since it is
> aligned with my interests, knowledge and professional goals.

Sounds like you already have a good background about some of the techniques you
might need for the project. I'd be interested to hear more about what you
already learned from courses, lectures, etc. if you'd like to elaborate. Also, I
think GSoC is great opertunity to get your hands dirty and to work on something
that you find really interessting.

> More precisely, the project named "Reinforcement Learning" caught my attention,
> since it combines the deep neural networks and the reinforcement learning. I
> already have some insights of the mlpack code base, because I started
> implementing the Gaussian Processes, which is still in my to do list. I read the
> project description very carefully together with the provided link, which I
> think is sufficient to get the idea and sense about the project. If there is
> something else to know, I would appreciate if you share it with me.

To be successful at this project, you should have a good knowledge of
reinforcement learning; i.e., you should be familiar with the way agents are
typically built and trained, and certainly, you should be familiar with the
individual components that you plan to implement. Besides that, especially for
the reinforcement learning project, there are some simple ways, to already dive
into the codebase and to learning something about the basics. Simply by
implementing a simple method like Policy Gradients that is capable to solve a
bunch of different tasks. Don't feel obligated, it's just an idea, we know it's
often not that easy to find the time to work on a project if you have a trillion
other things to do.

I hope this is helpful, let us know if you have any more questions.

Thanks,
Marcus

> On 27 Feb 2017, at 23:26, Vladimir Ilievski <ilievski.vladimir at live.com> wrote:
> 
> Hello everyone,
> 
> first of all I would like to congratulate to mlpack for being accepted as a participating organization in
> the Google Summer of Code 2017.
> 
> Since I am interested to take part in this year Google Summer of Code, let me introduce myself.
> My name is Vladimir and I am a Master Student at the Swiss Federal Institute of Technology in Lausanne (EPFL).
> I am pursuing my Masters in the field of Machine Learning and Intelligence. My particular areas of interest are
> Neural Networks, Reinforcement Learning and Gaussian Processes. Throughout the studies I followed several
> courses only related to these topics. Therefore, I have a solid knowledge about some principles in these areas.
> 
> However, I never had a chance to implement a solution for some of the concepts that requires analytical thinking
> and to write solid amount of code. Therefore, I would like to take part in the development of mlpack during GSoC,
> since it is aligned with my interests, knowledge and professional goals.
> 
> More precisely, the project named "Reinforcement Learning" caught my attention, since it combines the deep neural
> networks and the reinforcement learning. I already have some insights of the mlpack code base, because I started
> implementing the Gaussian Processes, which is still in my to do list. I read the project description very carefully together
> with the provided link, which I think is sufficient to get the idea and sense about the project. If there is something else to
> know, I would appreciate if you share it with me.
> 
> Best Regards,
> and expect my application :),
> 
> Vladimir
> _______________________________________________
> mlpack mailing list
> mlpack at lists.mlpack.org <mailto:mlpack at lists.mlpack.org>
> http://knife.lugatgt.org/cgi-bin/mailman/listinfo/mlpack <http://knife.lugatgt.org/cgi-bin/mailman/listinfo/mlpack>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://knife.lugatgt.org/pipermail/mlpack/attachments/20170228/dca26df0/attachment-0001.html>