07 max flow
TRANSCRIPT
-
8/6/2019 07 Max Flow
1/34
1
Chapter 7
Network Flow
Slides by Kevin Wayne.Copyright 2005 Pearson-Addison Wesley.All rights reserved.
-
8/6/2019 07 Max Flow
2/34
2
Soviet Rail Network, 1955
Reference: On the history of the transportation and maximum flow problems.Alexander Schrijver in Math Programming, 91: 3, 2002.
-
8/6/2019 07 Max Flow
3/34
3
Maximum Flow and Minimum Cut
Max flow and min cut.
Two very rich algorithmic problems.
Cornerstone problems in combinatorial optimization.
Beautiful mathematical duality.
Nontrivial applications / reductions.
Data mining.
Open-pit mining.
Project selection.
Airline scheduling.
Bipartite matching. Baseball elimination.
Image segmentation.
Network connectivity.
Network reliability.
Distributed computing.
Egalitarian stable matching.
Security of statistical data.
Network intrusion detection. Multi-camera scene reconstruction.
Many many more
-
8/6/2019 07 Max Flow
4/34
4
Flow network.
Abstraction for material flowing through the edges.
G = (V, E) = directed graph, no parallel edges.
Two distinguished nodes: s = source, t = sink.
c(e) = capacity of edge e.
Minimum Cut Problem
s
2
3
4
5
6
7
t
15
5
30
15
10
8
15
9
6 10
10
10154
4
capacity
source sink
-
8/6/2019 07 Max Flow
5/34
5
Def. An s-t cut is a partition (A, B) of V with s A and t B.
Def. The capacity of a cut (A, B) is:
Cuts
s
2
3
4
5
6
7
t
15
5
30
15
10
8
15
9
6 10
10
10154
4
Capacity = 10 + 5 + 15= 30
A
cap(A,B) = c(e)
e out ofA
"
-
8/6/2019 07 Max Flow
6/34
6
s
2
3
4
5
6
7
t
15
5
30
15
10
8
15
9
6 10
10
10154
4A
Cuts
Def. An s-t cut is a partition (A, B) of V with s A and t B.
Def. The capacity of a cut (A, B) is:cap(A,B) = c(e)
e out ofA
"
Capacity = 9 + 15 + 8 + 30= 62
-
8/6/2019 07 Max Flow
7/34
7
Min s-t cut problem. Find an s-t cut of minimum capacity.
Minimum Cut Problem
s
2
3
4
5
6
7
t
15
5
30
15
10
8
15
9
6 10
10
10154
4A
Capacity = 10 + 8 + 10= 28
-
8/6/2019 07 Max Flow
8/34
8
Def. An s-t flow is a function that satisfies:
For each e E: [capacity]
For each v V {s, t}: [conservation]
Def. The value of a flow f is:
Flows
4
0
0
0
0 0
0 4 4
0
0
0
Value = 40
f(e)
e in to v
" = f(e)e out ofv
"0 " f(e) " c(e)
capacity
flow
s
2
3
4
5
6
7
t
15
5
30
15
10
8
15
9
6 10
10
10154
4 0
v(f) = f(e)e out ofs
" .
4
-
8/6/2019 07 Max Flow
9/34
9
Def. An s-t flow is a function that satisfies:
For each e E: [capacity]
For each v V {s, t}: [conservation]
Def. The value of a flow f is:
Flows
10
6
6
11
1 10
3 8 8
0
0
0
11
capacity
flow
s
2
3
4
5
6
7
t
15
5
30
15
10
8
15
9
6 10
10
10154
4 0
Value = 24
f(e)
e in to v
" = f(e)e out ofv
"0 " f(e) " c(e)
v(f) = f(e)e out ofs
" .
4
-
8/6/2019 07 Max Flow
10/34
10
Max flow problem. Find s-t flow of maximum value.
Maximum Flow Problem
10
9
9
14
4 10
4 8 9
1
0 0
0
14
capacity
flow
s
2
3
4
5
6
7
t
15
5
30
15
10
8
15
9
6 10
10
10154
4 0
Value = 28
-
8/6/2019 07 Max Flow
11/34
11
Flow value lemma. Let f be any flow, and let (A, B) be any s-t cut.
Then, the net flow sent across the cut is equal to the amount leaving s.
Flows and Cuts
10
6
6
11
1 10
3 8 8
0
0
0
11
s
2
3
4
5
6
7
t
15
5
30
15
10
8
15
9
6 10
10
10154
4 0
Value = 24
f(e)e out ofA
" # f(e)e in to A
" = v( f)
4
A
-
8/6/2019 07 Max Flow
12/34
12
Flow value lemma. Let f be any flow, and let (A, B) be any s-t cut.
Then, the net flow sent across the cut is equal to the amount leaving s.
Flows and Cuts
10
6
6
1 10
3 8 8
0
0
0
11
s
2
3
4
5
6
7
t
15
5
30
15
10
8
15
9
6 10
10
10154
4 0
f(e)e out ofA
" # f(e)e in to A
" = v( f)
Value = 6 + 0 + 8 - 1 + 11= 24
4
11
A
-
8/6/2019 07 Max Flow
13/34
13
Flow value lemma. Let f be any flow, and let (A, B) be any s-t cut.
Then, the net flow sent across the cut is equal to the amount leaving s.
Flows and Cuts
10
6
6
11
1 10
3 8 8
0
0
0
11
s
2
3
4
5
6
7
t
15
5
30
15
10
8
15
9
6 10
10
10154
4 0
f(e)e out ofA
" # f(e)e in to A
" = v( f)
Value = 10 - 4 + 8 - 0 + 10= 24
4
A
-
8/6/2019 07 Max Flow
14/34
14
Flows and Cuts
Flow value lemma. Let f be any flow, and let (A, B) be any s-t cut. Then
Pf.
f(e)
e out ofA
" # f(e)= v(f)e in toA
" .
v(f) = f(e)e out ofs
"
=
v #A
" f(e)e out ofv
" $ f(e)e in to v
"%
&'
(
)*
= f(e)e out ofA
" $ f(e).e in to A
"
by flow conservation, all termsexcept v = s are 0
-
8/6/2019 07 Max Flow
15/34
15
Flows and Cuts
Weak duality. Let f be any flow, and let (A, B) be any s-t cut. Then the
value of the flow is at most the capacity of the cut.
Cut capacity = 30 Flow value 30
s
2
3
4
5
6
7
t
15
5
30
15
10
8
15
9
6 10
10
10154
4
Capacity = 30
A
-
8/6/2019 07 Max Flow
16/34
16
Weak duality. Let f be any flow. Then, for any s-t cut (A, B) we have
v(f) cap(A, B).
Pf.
Flows and Cuts
v( f) = f(e)e out ofA
" # f(e)e in to A
"
$ f(e)e out ofA
"
$ c(e)e out ofA
"
= cap(A,B)
s
t
A B
7
6
8
4
-
8/6/2019 07 Max Flow
17/34
17
Certificate of Optimality
Corollary. Let f be any flow, and let (A, B) be any cut.
If v(f) = cap(A, B), then f is a max flow and (A, B) is a min cut.
Value of flow = 28Cut capacity = 28 Flow value 28
10
9
9
14
4 10
4 8 9
1
0 0
0
14
s
2
3
4
5
6
7
t
15
5
30
15
10
8
15
9
6 10
10
10154
4 0A
-
8/6/2019 07 Max Flow
18/34
18
Towards a Max Flow Algorithm
Greedy algorithm.
Start with f(e) = 0 for all edge e E.
Find an s-t path P where each edge has f(e) < c(e).
Augment flow along path P.
Repeat until you get stuck.
s
1
2
t
10
10
0 0
0 0
0
20
20
30
Flow value = 0
-
8/6/2019 07 Max Flow
19/34
19
Towards a Max Flow Algorithm
Greedy algorithm.
Start with f(e) = 0 for all edge e E.
Find an s-t path P where each edge has f(e) < c(e).
Augment flow along path P.
Repeat until you get stuck.
s
1
2
t
20
Flow value = 20
10
10 20
30
0 0
0 0
0
X
X
X
20
20
20
-
8/6/2019 07 Max Flow
20/34
20
Towards a Max Flow Algorithm
Greedy algorithm.
Start with f(e) = 0 for all edge e E.
Find an s-t path P where each edge has f(e) < c(e).
Augment flow along path P.
Repeat until you get stuck.
greedy = 20
s
1
2
t
20 10
10 20
30
20 0
0
20
20
opt = 30
s
1
2
t
20 10
10 20
30
20 10
10
10
20
locally optimality global optimality
-
8/6/2019 07 Max Flow
21/34
21
Residual Graph
Original edge: e = (u, v) E.
Flow f(e), capacity c(e).
Residual edge.
"Undo" flow sent. e = (u, v) and eR = (v, u).
Residual capacity:
Residual graph: Gf = (V, Ef ).
Residual edges with positive residual capacity.
Ef = {e : f(e) < c(e)} {eR : f(e) > 0}.
u v17
6
capacity
u v11
residual capacity
6
residual capacity
flow
cf (e) =c(e)" f(e) if e # E
f(e) if eR # E
$%&
-
8/6/2019 07 Max Flow
22/34
22
Ford-Fulkerson Algorithm
s
2
3
4
5 t10
10
9
8
4
10
1062
G:capacity
-
8/6/2019 07 Max Flow
23/34
23
Augmenting Path Algorithm
Augment(f, c, P) {
b bottleneck(P)foreach e P {
if (e E) f(e) f(e) + b else f(eR) f(eR) - b
}
return f
}
Ford-Fulkerson(G, s, t, c) {
foreach e E f(e) 0Gf residual graph
while (there exists augmenting path P) {
f Augment(f, c, P)update Gf
}
return f
}
forward edge
reverse edge
-
8/6/2019 07 Max Flow
24/34
24
Max-Flow Min-Cut Theorem
Augmenting path theorem. Flow f is a max flow iff there are no
augmenting paths.
Max-flow min-cut theorem. [Elias-Feinstein-Shannon 1956, Ford-Fulkerson 1956]
The value of the max flow is equal to the value of the min cut.
Pf. We prove both simultaneously by showing TFAE:(i) There exists a cut (A, B) such that v(f) = cap(A, B).
(ii) Flow f is a max flow.
(iii) There is no augmenting path relative to f.
(i) (ii) This was the corollary to weak duality lemma.
(ii) (iii) We show contrapositive.
Let f be a flow. If there exists an augmenting path, then we can
improve f by sending flow along path.
-
8/6/2019 07 Max Flow
25/34
25
Proof of Max-Flow Min-Cut Theorem
(iii) (i)
Let f be a flow with no augmenting paths.
Let A be set of vertices reachable from s in residual graph.
By definition of A, s A.
By definition of f, t A.
v( f) = f(e)e out ofA
" # f(e)e in to A
"
= c(e)e out ofA
"
= cap(A,B)
original network
s
t
A B
-
8/6/2019 07 Max Flow
26/34
26
Running Time
Assumption. All capacities are integers between 1 and C.
Invariant. Every flow value f(e) and every residual capacity cf (e)
remains an integer throughout the algorithm.
Theorem. The algorithm terminates in at most v(f*) nC iterations.
Pf. Each augmentation increase value by at least 1.
Corollary. If C = 1, Ford-Fulkerson runs in O(mn) time.
Integrality theorem. If all capacities are integers, then there exists a
max flow f for which every flow value f(e) is an integer.Pf. Since algorithm terminates, theorem follows from invariant.
-
8/6/2019 07 Max Flow
27/34
7.3 Choosing Good Augmenting Paths
-
8/6/2019 07 Max Flow
28/34
28
Ford-Fulkerson: Exponential Number of Augmentations
Q. Is generic Ford-Fulkerson algorithm polynomial in input size?
A. No. If max capacity is C, then algorithm can take C iterations.
s
1
2
t
C
C
0 0
0 0
0
C
C
1 s
1
2
t
C
C
1
0 0
0 0
0X 1
C
C
X
X
X
1
1
1
X
X
1
1X
X
X
1
0
1
m, n, and log C
-
8/6/2019 07 Max Flow
29/34
29
Choosing Good Augmenting Paths
Use care when selecting augmenting paths.
Some choices lead to exponential algorithms.
Clever choices lead to polynomial algorithms.
If capacities are irrational, algorithm not guaranteed to terminate!
Goal: choose augmenting paths so that: Can find augmenting paths efficiently.
Few iterations.
Choose augmenting paths with: [Edmonds-Karp 1972, Dinitz 1970]
Max bottleneck capacity. Sufficiently large bottleneck capacity.
Fewest number of edges.
-
8/6/2019 07 Max Flow
30/34
30
Capacity Scaling
Intuition. Choosing path with highest bottleneck capacity increases
flow by max possible amount.
Don't worry about finding exact highest bottleneck path.
Maintain scaling parameter .
Let Gf () be the subgraph of the residual graph consisting of only
arcs with capacity at least .
110
s
4
2
t1
170
102
122
Gf
110
s
4
2
t
170
102
122
Gf (100)
-
8/6/2019 07 Max Flow
31/34
31
Capacity Scaling
Scaling-Max-Flow(G, s, t, c) {
foreach e E f(e) 0 smallest power of 2 greater than or equal to C
Gf residual graph
while ( 1) {Gf() -residual graph
while (there exists augmenting path P in Gf()) {f augment(f, c, P)update Gf()
}
/ 2 }
return f
}
-
8/6/2019 07 Max Flow
32/34
32
Capacity Scaling: Correctness
Assumption. All edge capacities are integers between 1 and C.
Integrality invariant. All flow and residual capacity values are integral.
Correctness. If the algorithm terminates, then f is a max flow.
Pf.
By integrality invariant, when = 1 Gf() = Gf. Upon termination of = 1 phase, there are no augmenting paths.
-
8/6/2019 07 Max Flow
33/34
33
Capacity Scaling: Running Time
Lemma 1. The outer while loop repeats 1 + log2 C times.
Pf. Initially C < 2C. decreases by a factor of 2 each iteration.
Lemma 2. Let f be the flow at the end of a -scaling phase. Then the
value of the maximum flow is at most v(f) + m .
Lemma 3. There are at most 2m augmentations per scaling phase. Let f be the flow at the end of the previous scaling phase.
L2 v(f*) v(f) + m (2).
Each augmentation in a -phase increases v(f) by at least .
Theorem. The scaling max-flow algorithm finds a max flow in O(m log C)augmentations. It can be implemented to run in O(m2 log C) time.
proof on next slide
-
8/6/2019 07 Max Flow
34/34
34
Capacity Scaling: Running Time
Lemma 2. Let f be the flow at the end of a -scaling phase. Then value
of the maximum flow is at most v(f) + m .
Pf. (almost identical to proof of max-flow min-cut theorem)
We show that at the end of a -phase, there exists a cut (A, B)
such that cap(A, B) v(f) + m .
Choose A to be the set of nodes reachable from s in Gf().
By definition of A, s A. By definition of f, t A.
v( f) = f(e)e out ofA
" # f(e)e in to A
"
$ (c(e)e out ofA
" # %) # %e in to A
"
= c(e)e out ofA
" # %e out ofA
" # %e in to A
"
$ cap(A, B ) - m%
original network
s
t
A B