Solutions to Homework 7
March 7, 2015
Exercise 1 (Ex 7.4.1, page 301). Give algorithms to decide the following:
a) Is L(G) finite, for a given CFG G? Hint: Use the pumping lemma.
b) Does L(G) contain at least 100 strings, for a given CFG G?
Proof. (a) Convert G to Greibach normal form (without useless symbols),
that is, all productions are of the form A → aγ, where γ ∈ V ∗ . Build a
directed graph with vertices V ( = variables), and add an edge A → B if B
appears in γ. L(G) is finite if and only if the graph does not contain cycles.
(b) Apply (a) first, and assume without loss of generality G is already
in Greibach normal form. If L(G) is infinite, return YES. Otherwise, enu-
meration all possible derivations from S; if there are at least 100 different
strings, return YES; otherwise, return NO.
Exercise 2 (Ex 7.4.2, page 308). Develop linear-time algorithms for the
following questions about CFG’s:
a) Which symbols appear in some sentential form?
b) Which symbols are nullable (derive ϵ)?
Proof. (a) Fix symbol a ∈ Σ. We want to know whether or not a appears
in some sentential form γ ∈ (V ∪ T )∗ . If suffices to know whether or not a
appears in each variable. Construct a graph with vertices V ∪ {a}. If there
is a production rule A → γ, where γ contains terminal a, then add an edge
A → a; If there is a production rule A → γ, where γ contains variable B,
then add an edge A → B. It is easy to check that a can appear in variable
A if and only if there is a path from A to a. Both the construction and
reachability test can be done in linear time.
1
(b) Initially, mark all variables A with production A → ϵ. If there is a
production A → B1 B2 . . . Bm , where all Bi ’s all nullable, then mark A as
nullable. This can be implemented in linear time (for each production of the
form A → B1 B2 . . . Bm , assign an integer denoting the number of nullable
variable on the right hand side, which is 0 initially; for each variable A, store
a list of pointers to all productions whose right hand side containing A, and
when A becomes nullable, update the corresponding counters).
Exercise 3 (Ex 8.2.2, page 335-336). Design Turing machines for the fol-
lowing languages:
a) The set of strings with an equal number of 0’s and 1’s.
b) {an bn cn : n ≥ 1}.
c) {wwR : w is any string of 0’s and 1’s}.
Proof. (a)
Figure 1: TM accepting strings with an equal number of 0’s and 1’s
(b) Please see lecture notes 13.
(c) See Figure 3.
Exercise 4 (Ex 8.2.3, page 336). Design a Turing machine that takes as
input a number N and adds 1 to it in binary. To be precise, the tape initially
contains a $ followed by N in binary. The tape head is initially scanning
the $ in state q0 . Your TM should halt with N + 1, in binary, on its tape,
scanning the leftmost symbol of N + 1, in state qf . You may destroy the $
2
Figure 2: TM accepting {wwR : w is any string of 0’s and 1’s}
in creating N + 1, if necessary. For instance, q0 $10011 ⊢∗ $qf 10100, and
q0 $11111 ⊢∗ qf 100000.
Figure 3: TM performing adding by one
Exercises are from the book “Automata Theory, Language, and Compu-
tation”, 3rd edition, by John E. Hopcroft, Rajeev Motwani, and Jeffrey D.
Ullman, published by Addison-Wesley.
