Most relevant graph algorithms for coding interviews.

Actually, an issue with this table:

Lazy Dijkstra takes O(V + E log E) time and O(V + E) space, while normal Dijkstra takes O(E log V) time and O(V) space.

The table shows the time for one and the space for the other.

Lazy dijkstra is the go-to version for interviews due to the lack of decrease-priority operation in most standard library heap implementations.

The same issue applies to Prim's algorithm for MST. It also has a lazy version with the same analysis.

The other classic MST algorithm, Kruskal, also takes O(V + E) space.

Also, to clarify, the space analysis in the table does not include constructing the adjacency list, which takes O(V + E) space.