To illustrate this phenomenon we consider a simple example of stochastic optimization problem andapplySGDwithconstant stepsizetosolveit.

Subsequently,wedefineequivariant energy based models to model invariant densities that are learned using contrastivedivergence.

(Algorithm 1). Then, we propose another strategy (Algorithm 2) to solve FSPCA for the general covariance by iteratively building a carefully designed proxy.