Avoiding for loops for multiplication of multidimensional matrices

I have to do a matrix multiplication of two vectors in principle to get a scalar. The vectors are (1 x 2) and (2 x 1). However, I need to do this for a lot of cases. Those cases are defined in matrix form like the following.

for i = 1:Niter
    for k = 1:Cn
        for ti = 1:T
            E(i, k, ti) = [D(ti, :) - squeeze(y(i, k, ti, :)).'] * [D(ti, :) - squeeze(y(i, k, ti, :)).'].'
        end
    end
end

So, `D` is a matrix with T x 2 size, y is a 4D matrix with size `Niter x Cn x T x 2`. Is there a way to avoid these for loops? I just want to compute the row and column vectors for all different cases `Niter`, `Cn` and `T`.

Answer 1

[Matt J]

D=reshape(D,1,1,T,2);

E=vecnorm(D-y,2,4).^2;

[Tworit Dash]

Thank you for the answer. Amazing. I think the 4th argument is 4, right? If I do 3, it does it in the 3rd axis. I still need T as the 3rd dimension for E. E should be [Niter x Cn x T].

[Matt J]

The 3rd argument to vecnomr should have been 4, correct.

[Tworit Dash]

Sorry yes, I meant the 3rd argument.