Matrix Factorization In Matlab using Stochastic Gradient Descent

I have to factorize Matrix R[m*n] to two low-rank Matrices (U[K*m] and V[K*n]), I do this for predicting missing values of R by U and V.

The problem is, for factorizing R I can't use Matlab factorization methods, so I have to work on objective function which minimizes the sum-of-squared-errors for enhancing factorization accuracy:

details are shown below:

My Question in this post is how to minimize function F in Matlab Using Stochastic Gradient Descent method to decompose R into U and V matrices.

[Matt J]

Since your function is not continuous/differentiable (because I_ij is not), I wonder whether any kind of gradient method applies.

How large are R, U, ad V typically. You might be able to use the genetic algorithm ga() in the Global Optimization Toolbox.