Set a X to the column vector, denoted x∈R^n, W=X*(X)'. But in CVX, the wrong expression , “Invalid quadratic form(s): not a square” is shown. How to modify ? Thank you!
That is non-convex,due to multiplying different variables. E.g., the (1,2) element of X*X' is X(1)*X(2), which is neither convex nor concave.
Depending on what is done with X*X', it is possible that the optimization problem is convex, and can be formulated in such as way as to never explicitly formulate X*X'. Or maybe it can’t. The readers don’t know, because we don’t know what your overall problem,is.
Please very carefully read
which you have previously been instructed to read.
Thank you! Now during a whole day with designning some CVX programs, significally making a great step for me! The concept about DCP ruleset understood is terrible , as if a clown is playing tricks on you!
Fortunately, I have grasped it and feel so happy! I like CVX, and same as detest it!
If you stay within its capabilities, CVX can be your best friend. If you stray outside these capabilities, it can be your cruelest mistress.