Recent content by slylock

  1. S

    Optimizing Linear Least Squares with Multiple Unknowns: A Partitioned Approach

    Why choose zeor for w and x. We have to minimize the equation for w and x and as far as i know the way was to take derivative with respect to each unknown vector and put it equal to zero and then solve the equation. This should give two normal equations. First take derivative with respect to x...
  2. S

    Optimizing Linear Least Squares with Multiple Unknowns: A Partitioned Approach

    Hmmm i don't understand what u mean here. Thing is forget about constants and then solve the equation. Assuming let's say there are no constants then how do you solve the equation. Technically it should give me two normal equations.
  3. S

    Optimizing Linear Least Squares with Multiple Unknowns: A Partitioned Approach

    x = v^{1} v^{2} ... v^{k} c = some constant values ... basically some x values we don't want to find or in other words want them to remain constant.
  4. S

    Optimizing Linear Least Squares with Multiple Unknowns: A Partitioned Approach

    Hi there, Could some one please tell me how can i solve this linear least square problem. x*,w* = argmin_{x,w}|| Gx - Mw ||^{2}_{2} subject to v_{k} = c_{k}; k = 1 ... p v_{k} is the vertex c is the constant Here x and w are the unknowns. G - is a 9m x 3n matrix - the...
Back
Top