Talk:Dijkstra's algorithm: Difference between revisions

Content deleted Content added
An interesting paper questioning optimality: They cite the earlier work on proof of universal optimality, and state that Dijkstra's algorithm is still optimal ''if you want to know the order of the points in the path''; their algorithm does not produce this ordering.
An interesting paper questioning optimality: It will be interesting to see if this gets out of pre-print status into WP:RS.
 
(One intermediate revision by the same user not shown)
Line 67:
 
== An interesting paper questioning optimality ==
This paper "Breaking the Sorting Barrier for Directed Single-Source Shortest Paths":
This paper:
 
https://arxiv.org/abs/2504.17033
Line 76:
 
They cite the earlier work on proof of universal optimality, and state that Dijkstra's algorithm is still optimal ''if you want to know the order of the points in the path''; their algorithm does not produce this ordering.
 
It will be interesting to see if this gets out of pre-print status into [[WP:RS]].
 
— [[User:The Anome|The Anome]] ([[User talk:The Anome|talk]]) 12:43, 1 June 2025 (UTC)