It is difficult to come up with a better way to start off the week than with news as good as this. As usual, we have two main types of project ideas in Shogun:
- Accessibility improvements.
- Core machine learning tasks or framework improvements.
Check out the full ideas list for more detailed descriptions.
Providing that Shogun has plenty of useful and interesting machine learning methods but, unfortunately, some of them are not so accessible for users that are not familiar with Shogun’s code base, this year the accessibility improvements projects seem to be particularly important. We expect to have after this summer more interactive demos showing off Shogun’s capabilities.
Nonetheless, there are also some interesting ideas concerning the implementation of new machine learning algorithms. For example, extensions in the Gaussian Processes to support classification, more dimension reduction techniques (are you an expert in ball trees? then we want you!) and some really challenging projects such as large-scale estimation of sparse Gaussian densities. Of course, last but not least, there is a very nice idea about my favourite topic: structured learning! This idea aims at providing some tools to target general structured output models with SO-SVMs, large-scale solvers and kernels.