It's hard to find reliable sources of information when starting out in deep learning. These will save you time, effort, and headaches. There are tons of free machine learning resources out there, but the sheer volume makes it difficult to sift through them. Not all of them are reliable or well-written. When you're starting out in deep learning, a poorly written tutorial can do more harm than good.

I was downvoted to hell on my main post, and the replies ("but Kaggle and Gaussian noise!") strongly suggest we have a lot of enthusiasts/amateurs in the subreddit. Working professionals attracting top dollar know that focusing on non-NN techniques is currently a poor idea at best. I can't convey that better than Google/Facebook/everyone else in the world already have, through their hiring and their solutions to pretty much everything in the past few years. I feel for the people new to the field, and they should have exposure (through books and basic problems) to other techniques. They should then take "big" data sets, run NNs and the other techniques on them, and see what the performance differences are.