notes on ICLR-2019 paper "Learning Deep Representations by Mutual Information Estimation and Maximization".
Momentum, Adagrad, Adadelta, RMSprop, Adam
How to speed up training pytorch model on GPU, avoid for looooop!
notes on reimplementation of InfoGraph.
Difference between nn.ModuleList and nn.Sequential in torch.
How sparse matrix is stored and applied in numpy, scipy, and torch.
Code notes on GCN pytorch.
Experiments about the MI estimator, MI maximization, etc.
Proof of the bias-variance tradeoff of the models.
Mutual Information Maximization concept, theory, application in Deep Learning.