Researchers at Ryerson University in Canada have recently published their work on Deep Super Learning. In short, super learning is an ensemble meant to find the best mix of learning algorithms:
"The deep super learner is flexible, adaptable, and easy to train with good performance across different tasks using identical hyper-parameter values" [source]
You can read more about the technical aspects of this in the release paper. It didn't take long until someone made an implementation of this concept in scikit-learn.
This repo is still in its infancy, so don't expect too much documentation just yet. But I'd suspect that if it catches and if it finds wide implementation within the field, documentation is not going to be a problem. And since we're talking about github and open-source, anyone can contribute to its development.
To stay in touch with me, follow @cristi
Cristi Vlad Self-Experimenter and Author
Very interesting. I will check out the paper. Thanks for sharing.
I took a very shallow look at the paper. Considering the proposed approach and methodology this would seem quite exciting. Although I am not in touch with the subject anymore but still retain the basics to get through the paper.
I remember my professor talking about "optimal combination of diverse learning algorithms" without any substantiating proof that the combination of this implementation is itself done by a higher algorithm. This was a couple of years ago.
Would love to get more info when the group publishes more
I'll be following so I'll probably post more about it.
Well as they said, only classification just as done in the paper, but it has a possibility for conversion into solving regression problems as well with some modifications But I'd suspect that if it catches and if it finds wide implementation within the field, documentation is not going to be a problem. And since we're talking about github and open-source, anyone can contribute to its development. i agree with this
Unfortunate on the lack of documentation aspect, makes it a little harder to dive into...you down to do a follow up on this with more resources? Would be interesting to have a brief tutorial on it.
Well, I might actually do that, hehe, once I understand the gist of it :)
excellent!
Your post is really great after I finished reading your article share I want to always connect with you and I follow you always.