I've been spending my free time getting myself familiar with ML and personally I'm focused on resharpening my math skills, reading papers from the "awesome deep learning papers" repo [1], playing with TensorFlow, and reading the Neural Networks and Deep Learning book [2]. I did my undergrad degree in math, so a lot of the math is just review for me, but ML seems to get fairly math heavy pretty quick. I would suggest anyone looking at ML to spend a good amount of time going through the backing math in addition to the CS parts, otherwise you might not develop much intuition for what's going on.
If you are serious about deep learning I would recommend this book instead: http://www.deeplearningbook.org It is coming out next month, but you can view the entire book online.
[1] https://github.com/terryum/awesome-deep-learning-papers
[2] http://neuralnetworksanddeeplearning.com/