Hello Prabhat, welcome and thanks for getting in touch.
> I'm also a member of my colleges Data Science group. We are a bunch of people > who organize open discussions and workshops related to various cool and > interesting topics of machine learning. Some of our members have also written > some awesome blogs. Really interesting blog posts, especially Singular Value Decomposition. Elucidated; mlpack implements various decomposition methods like svd, randomized svd, block Krylov, quic svd. > Presently, I'm working with Prof. Kusum Deep( google scholar) on convex > optimization. She is an expert on Nature inspired optimization techniques > especially Particle swarm optimization. She has also written a book on it . > Lately, I've been helping her in organizing a workshop on Particle swarm > optimization jointly organized by Soft Computing Research Society. That is great, are you familiar with PSO or did you already worked on a specific idea? > So, I'm very much interested in the Particle Swarm optimization project. I > would > like to know what can I do prior to Gsoc proposal submission regarding the > project. I can make a base PR for PSO like the one which is open or I can > implement some other optimizers for constrained optimization problems. The first step is to get familiar with the optimization framework; arxiv.org/abs/1711.06581 and http://www.mlpack.org/docs/mlpack- git/doxygen/optimizertutorial.html should be helpful. As you already pointed out there is an open PR for the standard PSO method, so if you like you can look into some improvements like MeanPSO ("Mean particle swarm optimisation for function optimisation" by K. Deep et al.) or DPSO ("The Improvement of Particle Swarm Optimization" by Z. Zhou et al.) comes to mind, but don't feel obligated. Let me know if I should clarify anything. Thanks, Marcus > On 2. Mar 2018, at 11:21, Prabhat Sharma <prabhatsharma7...@gmail.com> wrote: > > Hello everyone, > My name is Prabhat Sharma ( Github: Prabhat-IIT) and I am a sophomore at IIT > Roorkee, India pursuing Electrical Engineering. I am a machine learning > enthusiast and am well versed with C/C++, Python and Matlab. I've completed > some basic Machine Learning MOOCs and in the process implemented some pretty > famous algorithms from scratch. > Recently, I've Interned with Shopclues <https://www.shopclues.com/> an > e-commerce giant in India. I won a hackathon organized by the company and > helped them to build a merchant rating system for over 6 lakh merchants by > analysing real time data and feeding it to simple but effectively designed > machine learning algorithm which was formulated by me and my team members. > > I'm also a member of my colleges Data Science group > <https://github.com/dsgiitr>. We are a bunch of people who organize open > discussions and workshops related to various cool and interesting topics of > machine learning. Some of our members have also written some awesome blogs > <https://medium.com/data-science-group-iitr>. > > Presently, I'm working with Prof. Kusum Deep( google scholar > <http://scholar.google.com/citations?user=ByQN_c0AAAAJ>) on convex > optimization. She is an expert on Nature inspired optimization techniques > especially Particle swarm optimization. She has also written a book on it > <http://www.newagepublishers.com/servlet/nagetbiblio?bno=001699> . Lately, > I've been helping her in organizing a workshop <http://scrs.in/seta/> on > Particle swarm optimization jointly organized by Soft Computing Research > Society. > > So, I'm very much interested in the Particle Swarm optimization project. I > would like to know what can I do prior to Gsoc proposal submission regarding > the project. I can make a base PR for PSO like the one which is open or I can > implement some other optimizers for constrained optimization problems. > > Thanks and Regards > Prabhat Sharma > Sophomore, IIT Roorkee > _______________________________________________ > mlpack mailing list > mlpack@lists.mlpack.org > http://knife.lugatgt.org/cgi-bin/mailman/listinfo/mlpack
_______________________________________________ mlpack mailing list mlpack@lists.mlpack.org http://knife.lugatgt.org/cgi-bin/mailman/listinfo/mlpack
