Undecidable problems

for
Recursively enumerable languages

continued…

1
Take a recursively enumerable language L

Decision problems:
• L is empty?

• L is finite?
• L contains two different strings
of the same length?

All these problems are undecidable

2
Theorem:
For a recursively enumerable language L
it is undecidable to determine whether
L is finite

Proof:
We will reduce the halting problem
to this problem

3
 Let M be the TM with L( M ) = L

Suppose we have a decider

for the finite language problem:

YES L(M ) finite

finite language
M problem
decider NO L(M ) not finite

4
 We will build a decider
for the halting problem:

M YES M halts on w
Halting problem
decider
w NO doesn’t
M w
halt on

5
 We want to reduce the halting problem to
the finite language problem

Halting problem decider

YES NO
M finite language
Mw
problem
w YES
decider NO

6
 We need to convert one problem instance
to the other problem instance

Halting problem decider

YES NO
M convert finite language
Mw
input problem
w YES
? decider NO

7
Construct machine M w:
On arbitrary input string s

Initially, simulates M on input w

If M enters a halt state,

accept s ( Σ* inifinite language)

Otherwise, reject s ( ∅ finite language)

8
 M halts on w
if and
only if

L( M w ) is infinite

*
L( M w ) = Σ

9
 halting problem decider

YES NO
M construct M w finite language
problem
w Mw decider YES
NO

10
Take a recursively enumerable language L

Decision problems:
• L is empty?

• L is finite?
• L contains two different strings
of the same length?

All these problems are undecidable

11
Theorem:

For a recursively enumerable language L

it is undecidable to determine whether
L contains two different strings of
same length

Proof: We will reduce the halting problem

to this problem

12
Let M be the TM with L( M ) = L

Suppose we have the decider

for the two-strings problem:

YES L(M ) contains

Two-strings
M problem
decider NO Doesn’t
L(M )
contain
two equal length strings
13
 We will build a decider for
the halting problem:

M YES M halts on w
Halting problem
decider
w NO doesn’t
M w
halt on

14
 We want to reduce the halting problem to
the empty language problem

Halting problem decider

YES YES
M M w Two-strings
problem
w decider
NO NO

15
 We need to convert one problem instance
to the other problem instance

Halting problem decider

YES YES
M convert M Two-strings
w
inputs problem
w decider
NO NO
?

16
Construct machine Mw :
On arbitrary input string s
Initially, simulate M on input w

When M enters a halt state,

accept if s = a or s = b
(two equal length strings L( M w ) = {a, b} )

Otherwise, reject s ( L( M w ) = ∅ )
17
 M halts on w
if and
only if

M w accepts two equal length strings

Mw accepts a and b
18
 Halting problem decider

YES YES
M construct M w Two-strings
problem
w Mw decider
NO NO

19
Rice’s Theorem

20
Definition:

Non-trivial properties of
recursively enumerable languages:

any property possessed by some (not all)

recursively enumerable languages

21
Some non-trivial properties of
recursively enumerable languages:

• L is empty

• L is finite

• L contains two different strings

of the same length

22
Rice’s Theorem:
Any non-trivial property of
a recursively enumerable language
is undecidable

23
The Post Correspondence
Problem

24
Some undecidable problems for
context-free languages:

• Is L(G1 ) ∩ L(G2 ) = ∅ ?
G1,G2 are context-free grammars

• Is context-free grammar G ambiguous?

25
We need a tool to prove that the previous
problems for context-free languages
are undecidable:

The Post Correspondence Problem

26
The Post Correspondence Problem

Input: Two sequences of n strings

A = w1, w2 ,  , wn

B = v1, v2 ,  , vn

27
There is a Post Correspondence Solution
if there is a sequence i, j , , k such that:

PC-solution: wi w j  wk = vi v j vk

Indices may be repeated or omitted

28
Example: w1 w2 w3
A: 100 11 111

v1 v2 v3
B: 001 111 11

PC-solution: 2,1,3 w2 w1w3 = v2v1v3

11100111
29
 Example: w1 w2 w3
A: 00 001 1000

v1 v2 v3
B: 0 11 011

There is no solution

Because total length of strings from B

is smaller than total length of strings from A
30
The Modified Post Correspondence Problem

Inputs: A = w1, w2 ,  , wn

B = v1, v2 ,  , vn

MPC-solution: 1, i, j , , k

w1wi w j  wk = v1vi v j vk

31
Example: w1 w2 w3
A: 11 111 100

v1 v2 v3
B: 111 11 001

MPC-solution: 1,3,2 w1w3w2 = v1v3v2

11100111
32
Theorems

1. The MPC problem is undecidable

(by reducing the membership to MPC)

2. The PC problem is undecidable

(by reducing MPC to PC)

33
Theorem: The MPC problem is undecidable

Proof: We will reduce the membership

problem to the MPC problem

34
Membership problem

Input: recursive language L

string w

Question: w∈ L ?

Undecidable

35
Membership problem

Input: unrestricted grammar G

string w

Question: w∈ L(G ) ?

Undecidable

36
 Suppose we have a decider for
the MPC problem

String Sequences MPC solution?

A YES
MPC problem
decider
B NO

37
We will build a decider for
the membership problem

w∈ L(G ) ?
G Membership YES
problem
w decider NO

38
 The reduction of the membership problem
to the MPC problem:

Membership problem decider

A yes yes
G MPC problem
B decider no no
w

39
We need to convert the input instance of
one problem to the other

Membership problem decider

A yes yes
G convert MPC problem
inputs decider no no
w B
?

40
 A B Grammar G

FS ⇒ F S : start variable
F : special symbol

a a For every symbol a

V V For every variable V

41
 A B Grammar G

E ⇒ wE string w
E : special symbol

y x For every production

x→ y

⇒ ⇒

42
Example:

Grammar G : S → aABb | Bbb

Bb → C
AC → aac

String w = aaac

43
 A B

w1 : FS ⇒ v1 : F
w2 : a v2 : a
b b
c c
 A  A
B B
C C
w8 : S v8 : S 44
 A B

w9 : E⇒ v9 : aaacE
aABb S
Bbb S
 
C Bb
aac AC
w14 : ⇒ v14 : ⇒

45
Grammar G : S → aABb | Bbb
Bb → C
AC → aac

aaac ∈ L(G )

S ⇒ aABb ⇒ aAC ⇒ aaac

46
 S ⇒ aABb

A
w1 w10

F S ⇒a A B b

v1 v10

B
47
 S ⇒ aABb ⇒ aAC

A
w1 w10 w14 w2 w5 w12

F S ⇒a A B b⇒a A C

v1 v10 v14 v2 v5 v12

48
 S ⇒ aABb ⇒ aAC ⇒ aaac

A
w1 w10 w14 w2 w5 w12w14 w2 w13 w9

F S ⇒a A B b⇒a A C ⇒a a a c E

v1 v10 v14 v2 v5 v12 v14 v2 v13 v9

49
( A, B ) has an MPC-solution

if and
only if

w∈ L(G )

50
 Membership problem decider

A yes yes
G Construct MPC problem
A, B B decider no no
w

51
Since the membership problem is undecidable,
The MPC problem is undecidable

END OF PROOF
52
Theorem: The PC problem is undecidable

Proof: (omitted) The MPC problem can be

reduced to the PC problem

53
Some undecidable problems for
context-free languages:

• Is L(G1 ) ∩ L(G2 ) = ∅ ?
G1,G2 are context-free grammars

• Is context-free grammar G
ambiguous?

We reduce the PC problem to these problems

54