Here's a good book on convolutional neural networks: http://www.deeplearningbook.org/ It's suppose to be like the bible on deep learning at the moment but like most cutting edge stuff -- you're going to have to put a lot of effort into thinking for yourself and reading a lot of papers to make any new progress.

I'll hold off until machine learning learned itself a course in machine learning to teach me machine learning to a level i can give courses in machine learning to machines that give machine learning courses, some how that sounds more appealing :)

This seems like a really good resource but as another poster mentioned, it is unfinished and going from the GitHub there hasn't been any work done on it since December 2015. Of course, the developer may well return to it or a fork may overtake it. Let's see.

https://github.com/hal3/ciml/graphs/contributors?from=2015-0...

Past discussions https://news.ycombinator.com/from?site=ciml.info

Very nice work. If the eBook is reasonably priced when the book is done, I will buy a copy to support the author. I have read, or at least skimmed through, many ML books, and it is interesting to see how different authors think about and organize the material, and different perspectives help. I make the same comment on the ML and data science classes I have taken online: I like to get different perspectives on the same material from different teachers.

half the chapters are empty...

The link to "purchasing a ($55) print copy" actually links back to the same page.

If you have time to read this material and play / work with this, you're so lucky.