Adagrad
- PDF paper on Adaptive Subgradient Methods: http://www.jmlr.org/papers/volume12/duchi11a/duchi11a.pdf
- Notes on AdaGrad [PDF]: http://seed.ucsd.edu/
- https://xcorr.net/ : eliminating learning rates in stochastic gradient descent
- What is the purpose of AdaGrad? Quora