Hello developers, I am Shashank Shekhar, a junior at Indian School of Mines. I would like to contribute to Shogun under GSOC 2016. I am strongly interested in neural nets and hence would like to work on the 'Hip Deep Learning' project. I have some background knowledge on the topic and have done my own implementations of feedforward nets, CNNs, autoencoders in the past. I also have a knowledge of specific optimization methods like momentum, adadelta, adagrad and regularization methods like dropout etc. for deep nets.
However while going through the project requirements it was mentioned that knowledge of graphical models and optimization is a pre-requisite for it. I am more than willing to dive into literature for the project but as of now I have preliminary knowledge of optimization and know next to nothing about graphical models. Will this project be a good fit for me? In case you think I still have a shot at it, I would like to point out that most of the linked entrance tasks mentioned for the project already have merged or open PRs. The ones that don't have PRs yet are 'Add preprocessor layer for neural network' and 'Output gradient information in neural nets' which are from last year and I am not sure if someone is working on them. It would be great if you could point me to some entrance level tasks for this project. Regards, Shashank Shekhar
