I'm a bit confused about the complexity of FF since we didn't really specify it:
Should I assume that the FF complexity is the same as the Edmond-Karp implementation?
In some of the slides in the recitations we showed that FF performs better than Dinic.
(e.g: Tirgul 10 - slide 21)
Why would the FF complexity ever be better than Dinic?
(in the slide 21 example, if I use the EK complexity for FF it seems equal to Dinic.
Both algorithms create residual networks so it doesn't occur to me as if FF does less work than Dinic here)