This quantity is approximated via the convergence of iteratively reweighted least squares in Weiszfeld’s algorithm. In image processing, salt-and-pepper noise is effectively reduced by a nonlinear filter named for this quantity. Two subproblems have size one-fifth n and seven-tenths n in the recurrence of an algorithm that computes this quantity. Using two heaps computes this quantity for a stream input in a problem about its “maintenance.” This quantity appears twice in the name of an (*) pivot-finding algorithm that runs in asymptotically worst-case linear time. The optimal runtime of quicksort occurs when the pivot always equals this value of the subarray. (10[1])One way to find this quantity is to sort an array and get the element at index closest to half the array length. For 10 points, (10[1])name this quantity equal (10[1])to the 50th percentile of a data set. ■END■