18.404/6.

840 Intro to the Theory of Computation

Instructor: Mike Sipser

TAs:
- Fadi Atieh, Damian Barabonkov,
- Alex Dimitrakakis, Thomas Xiong,
- Abbas Zeitoun, and Emily Liu

1
 18.404 Course Outline

Computability Theory 1930s – 1950s

- What is computable… or not?
- Examples:
program verification, mathematical truth Complexity Theory 1960s – present
- Models of Computation: - What is computable in practice?
Finite automata, Turing machines, …
- Example: factoring problem
- P versus NP problem
- Measures of complexity: Time and Space
- Models: Probabilistic and Interactive computation

2
 Course Mechanics
Zoom Lectures
- Live and Interactive via Chat
- Live lectures are recorded for later viewing
Zoom Recitations
- Not recorded
- Two convert to in-person Homework bi-weekly – 35%
- Review concepts and more examples - More information to follow
- Optional unless you are having difficulty Midterm (15%) and Final exam (25%)
Participation can raise low grades
- Open book and notes
- Attend any recitation
Text Check-in quizzes for credit – 25%
- Introduction to the Theory of Computation - Distinct Live and Recorded versions
Sipser, 3rd Edition US. (Other editions ok but - Complete either one for credit within 48 hours
are missing some Exercises and Problems). - Initially ungraded; full credit for participation
3
 Course Expectations

Prerequisites
Prior substantial experience and comfort with
mathematical concepts, theorems, and proofs.
Creativity will be needed for psets and exams.

Collaboration policy on homework

- Allowed. But try problems yourself first.
- Write up your own solutions.
- No bibles or online materials.

4
 Role of Theory in Computer Science

1. Applications
2. Basic Research
3. Connections to other fields
4. What is the nature of computation?

5
 Let’s begin: Finite Automata
0 1
!1 0,1
1
*1 *2 *3
0

Input: finite string

Output: Accept or Reject
States: *1 *2 *3
1
Computation process: Begin at start state,
Transitions: read input symbols, follow corresponding transitions,
Accept if end with accept state, Reject if not.
Start state:
Examples: 01101 → Accept
Accept states: 00101 → Reject
!1 accepts exactly those strings in # where
# = {&| & contains substring 11}.

Say that # is the language of !1 and that !1 recognizes # and that # = -(!1).
6
 Finite Automata – Formal Definition

Defn: A finite automaton ! is a 5-tuple (#, Σ, &, '0, ))

# finite set of states
Σ finite set of alphabet symbols
& transition function &: #×Σ → # Example:
a
& (', .) = 0 means ' 0
!1
0
'0 start state 1
0,1
'1 '2 1 '3
) set of accept states
0

!1 = (#, Σ, &, '1, )) &= 0 1

# = {'1, '2, '3}
'1 '1 '2
Σ = {0, 1} '2 '1 '3
) = {'3} '3 '3 '3
7
 Finite Automata – Computation

Strings and languages

- A string is a finite sequence of symbols in Σ
- A language is a set of strings (finite or infinite)
- The empty string ε is the string of length 0
- The empty language ø is the set with no strings Recognizing languages
- :(#) = {$| # accepts $}
- :(#) is the language of #
Defn: # accepts string $ = $1$2 … $) each $* + Σ - # recognizes :(#)
if there is a sequence of states ,0, ,1, ,2, , … , ,) + /
where:
- ,0 = 00
- ,* = 1(,345 , $*) for 1 ≤ * ≤ ) Defn: A language is regular if some
- ,) + 8 finite automaton recognizes it.
8
 Regular Languages – Examples

0
"1 1
0,1
1
81 82 83
0

More examples:
! "# = {&| & contains substring 11} = 5
Let 6 = & & has an even number of 1s}
Therefore 5 is regular 6 is regular (make automaton for practice).

Let 7 = & & has equal numbers of 0s and 1s}

7 is not regular (we will prove).

Goal: Understand the regular languages

9
 Regular Expressions
Regular operations. Let !, # be languages:
- Union: ! ∪ # = & & ∈ ! or & ∈ #}
- Concatenation: ! ∘ # = *+ * ∈ ! and + ∈ #} = !#
- Star: !∗ = *- … */ each *0 ∈ ! for 1 ≥ 0}
Note: ε ∈ !∗ always Regular expressions
- Built from Σ, members Σ, ∅, ε [Atomic]
Example. Let ! = {good, bad} and # = {boy, girl}. - By using ∪,∘,∗ [Composite]

- ! ∪ # = {good, bad, boy, girl} Examples:

- ! ∘ # = !# = {goodboy, goodgirl, badboy, badgirl} - 0 ∪ 1 ∗ = Σ ∗ gives all strings over Σ
- !∗ = {ε, good, bad, goodgood, goodbad, badgood, - Σ ∗1 gives all strings that end with 1
badbad, goodgoodgood, goodgoodbad, … } - Σ ∗11Σ ∗ = all strings that contain 11 = : ;-

Goal: Show finite automata equivalent to regular expressions

10
 Closure Properties for Regular Languages
Theorem: If !", !$ are regular languages, so is !" ∪ !$ (closure under ∪)
Proof: Let &" = ()", Σ, +" , ," , -" ) recognize !"
&$ = ()$ , Σ, +$ , ,$ , -$ ) recognize !$
Construct & = () , Σ , + , ,0, - ) recognizing !" ∪ !$
& should accept input 0 if either &" or &$ accept 0.
Components of 2:
&Check-in
" 1.1
, ) = )" ×)$
In the proof, if &" and &$ are finite automata = ," , ,$ ," ∈ )" and ,$ ∈ )$ }
where &" has 8" states and &$ has 8$ states
,6 = (," , ,$ )
Then how many states does & have?
&$ (a) 8" + 8$ + ,, 1 , 7 = +" ,, 7 , +$ 1, 7
(b) 8" $ + 8$ $ - = -" ×-$ NO! [gives intersection]
(c) 8" ×8$ 1
- = -" ×)$ ∪ )" ×-$
Check-in 1.1
11
 Closure Properties continued
Theorem: If !", !$ are regular languages, so is !"!$ (closure under ∘)
Proof: Let &" = ()", Σ, +" , ," , -" ) recognize !"
&$ = ()$ , Σ, +$ , ,$ , -$ ) recognize !$
Construct & = () , Σ , + , ,0, - ) recognizing !" !$

&" &$

& should accept input 0

if 0 = 12 where
&
&" accepts 1 and &$ accepts 2.

0
1 2
Doesn’t work: Where to split 0?
12
 Quick review of today
1. Introduction, outline, mechanics, expectations
2. Finite Automata, formal definition, regular languages
3. Regular Operations and Regular Expressions
4. Proved: Class of regular languages is closed under ∪
5. Started: Closure under ∘ , to be continued…

13
MIT OpenCourseWare
https://ocw.mit.edu

18.404J / 18.4041J / 6.840J Theory of Computation

Fall 2020

For information about citing these materials or our Terms of Use, visit: https://ocw.mit.edu/terms.