Deep-Learning-Papers-Reading-Roadmap

Songrotek, GitHub, Oct 25, 2016
Commentary by Stephen Downes
files/images/coordinates.JPG

This is an unfinished work, but it illustrates nicely the use of academic papers as open educational resources by sequencing useful and important resources in such a way as to guide the reader through the essentials of a discipline. "The roadmap is constructed in accordance with the following four guidelines: from outline to detail; from old to state-of-the-art; from generic to specific areas (and) focus on state-of-the-art." It's best to think of this as a proto-MOOC. People can (and should) add resources (not just papers and books), and these can create branches and sub-branches. The resources themselves are all openly accessible. GitHub does provide limited social interaction, but you would expect a social network or community to grow around this collection. Actual MOOC classes would involve a self-managing cohort moving through the material together. Yes, it takes commitment and effort to learn a subject this way, and a lot of people don't have the skills. That's where educational institutions and student support should come in.

Views: 0 today, 369 total (since January 1, 2017).[Direct Link]