I am trying to solve problems using CVX. My code is
cvx_begin
variable k
variable D(m, n)
minimize(norm(k*eye(n)-D’*D, ‘fro’));
subject to
k>=1;
0<=D<=1;
cvx_end
I got the error: Only scalar quadratic forms can be specified in CVX. I know the problem is D’*D. But I don’t know how to rewrite it in DCP rules. Can anyone help me with that, thanks very much!