Why they’re worth studying: Radix heaps are interesting in that it’s possible to start off with a fairly
straightforward set of observations about how Dijkstra’s algorithm operates to get an O(m + nC)-time im-
plementation, and then refine that first to O(m + n log C) and from there to O(m + n (log C)
1/2
) through
more and more creative observations. In that sense, this is a fairly accessible data structure that goes
pretty deep into Dijkstra’s algorithm. Later algorithms and data structures have improved upon the run-
time even further, and this could also be a great launching point for further exploration.
20 / 22
Do'stlaringiz bilan baham: |