This repository includes algorithms in statistical computing and big data analysis.

For the first part, we derive the MLE for the most of the time while for the latter one, we minimize a loss function.So it may be a little diffence when describing an algorithm in these two scenarios. However, most of the time, we just need to add a minus sign to switch between these two situations, i.e. one is for minimization, the other is for maximization.

It can used in incomplete data set problem, see here.

See here.

This part mainly includes notes from Optimization Methods for Large-Scale Machine Learning by L´eon Bottou et al, see here

Push-Pull method, distributed stochastic gradient tracking(DSGT) method.

Can see here