DFA minimization

Example
• Construct a DFA over alphabet {0, 1} that
accepts those strings that end in 111 0
0 q
0 q 1
1
q q
1
0 q …

…
q
1 q
0 q
1
q …

…
1
q 1
q 1

• This is big, isn’t there a smaller DFA for this?

 Smaller DFA

0
1 1 1
q q q q 1
0
0

Can we do it with 3 states?

 Even smaller DFA?
• Suppose we had a 3 state DFA M for L

… let’s imagine what happens when:

M
inputs:

, 1, 11, 111

• By the pigeonhole principle, on two of these

inputs M ends in the same state
 Pigeonhole principle
Suppose you are tossing m balls into n bins, and
m > n. Then two balls end up in the same bin.

• Here, balls are inputs, bins are states:

If you have a DFA with n states and you run it on

m inputs, and m > n, then two inputs end up in
same state.
 A smaller DFA
• Suppose M ends up in the same state after
reading inputs x = 1 and y = 11
M 1, 11
inputs:

, 1, 11, 111 11, 111

“ends in 111”
• Then after reading one more 1
– The state of x1 = 11 should be rejecting
– The state of y1 = 111 should be accepting
… but they are both the same state!
 A smaller DFA
• Suppose M ends up in the same state after
reading inputs x =  and y = 1
M , 1
inputs:

, 1, 11, 111 11, 111

“ends in 111”
• Then after reading 11
– The state of x1 = 11 should be rejecting
– The state of y1 = 111 should be accepting
… but they are both the same state!
 No smaller DFA!
• After looking at all possible pairs for x, y, x ≠ y
(, 1) (, 11) (, 111) (1, 11) (1, 111) (11, 111)

we conclude that
There is no DFA with 3 states for L

• So, this DFA is minimal

0
1 1 1
q q q q 1
0
0

0
 DFA minimization 0

q  1
0 q 
1 1

…
q
q  … q 
1
1

…
0 1
q  … q
1 q 
1
q q  1
0 q
0 q  0
1
1 0
0 0
q
q 1 q

We will show how to turn any DFA for L into the

minimal DFA for L
 Minimal DFAs and distinguishable
states
• First, we have to understand minimal DFAs:
reject accept
0
1 1 1
q q q q 1
0
0

every pair of states

minimal DFA
is distinguishable
 Distinguishable states
• Two states q and q’ are distinguishable if
p t
w1 w2 wk-1 wk cce
… a
q

c t
w1 w2 wk-1 wk j e
… re
q’

on the same continuation string w1w2...wk, one

accepts, but the other rejects
Examples of distinguishable states
0 1 1

1 0 1
q q q q

0
0

(q0, q1) distinguishable by 01

(q0, q2) distinguishable by 1
(q0, q3) distinguishable by  DFA is minimal
(q1, q2) distinguishable by 1
(q1, q3) distinguishable by 
(q2, q3) distinguishable by 
Examples of distinguishable states
0 q q
q 0
0

0, 1

0, 1
q0 1
1
q 1
q 0, 1 q
0, 1

(q0, q3) distinguishable by 

(q1, q3) distinguishable by 
(q2, q3) distinguishable by  indistinguishable pairs
(q1, q2) distinguishable by 0 can be merged
(q0, q2) distinguishable by 0
(q0, q1) indistinguishable
Examples of distinguishable states
0
q q
0, 1 0, 1
q q
1 0, 1
q 0, 1 q

0, 1

(q0, q2) distinguishable by 

(q1, q2) distinguishable by 
(q0, q3) distinguishable by 
(q1, q3) distinguishable by 
(q0, q1) indistinguishable
(q2, q3) indistinguishable
Finding (in)distinguishable states
If q is accepting and q’ is rejecting
Rule 1: q x q’ Mark (q, q’) as distinguishable (x)

q1 x q1 ’

a a If (q1, q1’) are marked,

Rule 2: Mark (q2, q2’) as distinguishable (x)
q2 x q2 ’

Unmarked pairs are indistinguishable

Rule 3: Merge them together
 Example of DFA minimization
0
q
0 q
q q
0 1 1
q 0 q
q 1 0 q
1 0 q 1
q 0 q
1
q q
q q q q q q
1
 Example of DFA minimization
0
q
0 q
q q
0 1 1
q 0 q
q 1 0 q
1 0 q 1
q 0 q
1
q q x x x x x x
q q q q q q
1

 q11 is distinguishable from all other states

 Example of DFA minimization
0
q
0 q
q q x x
0 1 1
q 0 q x
q 1 0 q
1 0 q 1
q 0 q x
1
q q x x x x x x
q q q q q q
1

 q1 is distinguishable from q, q0, q00, q10

On transition 1, they go to distinguishable states
 Example of DFA minimization
0
q
0 q
q q x x
0 1 1
q 0 q x
q 1 0 q x x x
1 0 q 1
q 0 q x x
1
q q x x x x x x
q q q q q q
1

 q01 is distinguishable from q, q0, q00, q10

On transition 1, they go to distinguishable states
 Example of DFA minimization
0
q A
0 q
A q q x x
0 1 1
q 0 q A A x
q 1 0 q x x B x
1 0 q 1
q 0 q A A x A x
B 1
q q x x x x x x
C
q q q q q q
1

 Merge states not marked distinguishable

q, q0, q00, q10 are equivalent → group A
q1, q01 are equivalent → group B
q11 cannot be merged → group C
 Example of DFA minimization
0
q A
0 q
q q x x
A 0 1 1
q 0 q A A x
q 1 0 q x x B x
1 0 q 1
q 0 q A A x A x
B 1
q q x x x x x x
C q q q q q q
1

0
1 1
minimized DFA: q q qC 1
0
0
 Food for thought
• Why does method find all distinguishable pairs?

w1 w2 wk-1 wk
…
q
x x x x
w1 w2 wk-1 wk
…
q’

Because we work backwards

 Food for thought
• Why are there no inconsistencies when we merge?
B
w
q5
a

A q2
q3
q1
q7
a C
w
q6

Because we only merge indistinguishable states

 Food for thought
• Why is there no smaller DFA?
Suppose there is

By the pigeonhole principle this must happen:

M smaller DFA M’
v q
q’’
v, v’
v’ q’
 Food for thought
• Why is there no smaller DFA?
But then
M smaller DFA M’
v q w
q’’ w
v, v’
?
v’ q’ w

Every pair of states q’’ cannot exist!

is distinguishable