Shortest Path Problem: Difference between revisions

From Rice Wiki
No edit summary
No edit summary
Line 16: Line 16:
* Single pair problem: Shortest path between input pair
* Single pair problem: Shortest path between input pair


== Implementation ==
== Implementation: Bellman Ford ==


All shortest path must have <math> \leq n - 1 </math> edges. If this
All shortest path must have <math> \leq n - 1 </math> edges. If this
Line 22: Line 22:
therefore it is not the shortest.
therefore it is not the shortest.


<pre>
OPT(a, n-1) = min(w(u,a) + OPT(n-1, u) for all (u, a) in E)
OPT(a, n-1) = min(w(u,a) + OPT(n-1, u))
</pre>


where
<math> (u, a) \in E </math>
=== Bellman Ford Algorithm ===
Time complexity: <math> O(V^2) </math>
Time complexity: <math> O(V^2) </math>
<pre>
<pre>

Revision as of 01:49, 28 February 2024

Definitions

A path is a sequence of nodes such that for all consecutive nodes, there exist an edge

Let there be a weight assigned to each edge.

Single Source Shortest Path (SSSP)

Given a graph , source node , outupt the shortest path from the source

Variants

  • Single destination problem: shortest path from all nodes to a single destination
  • Single pair problem: Shortest path between input pair

Implementation: Bellman Ford

All shortest path must have edges. If this condition is not satisfied, there is a cycle in in the path, and therefore it is not the shortest.

OPT(a, n-1) = min(w(u,a) + OPT(n-1, u) for all (u, a) in E)

Time complexity:

for e = 1 to n - 1:
    for each u in V:
        OPT(u, e) = min(w(v,u) + OPT(v, e-1) for all (v, u))
        OPT(u, e) = min(OPT(u,e), OPT(u, e-1))