CMPSCI 601: Theory of Computation
David Mix Barrington
Spring, 2003
This is the home page for CMPSCI 601.
CMPSCI 601 is the graduate core course in
the theory of computation and will deal with formal
languages, computability, logic, and complexity theory.
Important Course Material:
Announcements (28 August 2003):
 I have moved the web site for the summer video version of the
course here, and started a new web site for the
fall version here.
 I am offering this course again during the summer term through
the PEEAS program, using videotaped
lectures from this Spring 2003 offering. This is
the main web page for the summer version of the course.
 The final exam has now been posted,
together with solutions.
 I have received and graded the final exams for the nine offcampus
students who took it. The scores, in descending order, were 68, 66, 65, 62,
57, 56, 54, 47, and 41. For overall course grades these students received
one AB, four B's, three BC's, and a C  they have been emailed with their
individual results.
 Here is a summary of the results of the final for the 39
oncampus students.
 The A range was 80+, AB range 6479, B range 4863, BC range
3247.
 There were seventeen A's, with the top scores being a 98, a 96,
a 93 and two 92's. There were eleven AB's, six B's, and five BC's.
 The course grades, with the midterm and homework factored in,
were generally higher than the final exam grades. There were 22 A's,
11 AB's, five B's, and an incomplete. As it happens, the five oncampus
B's were the same five who scored less than 50 points on the final exam.
 Solutions to HW#8 are now available
online.
 Oncampus finals are now graded. I will hand them back in person
in my office. I'll next be in today from around 1:30 to 5:00. Graded
homework will be available in the main office in the pickup box.
 There is a correction and a clarification to the crib sheet
included in the offcampus final exams:
 The language K, as it has bee all term, is defined to be {n: n is in
L(M_{n}), not the complement of this language which is Kbar.
 A more precise definition of the language VALCOMP_{M} is
{w: ∃ x: w describes an accepting computation of M on x}. Here
M is a onetape Turing machine fixed for the language.
 The Q&A page now has some questions relevant
to the final, some concerning the practice final and its solutions.
 The solution to the practice
final is now up.
 I have given the final exam to the PEEAS office, who will
mail it to offcampus students to arrive on Mon 19 May, just as the
oncampus students are taking theirs. I need the offcampus exams
back by Tue 27 May in order to file grades the next day. Offcampus
students may fax them that day or send them by overnight service to
arrive that day. Note that Mon 26 May is a national holiday.

 Paper copies of the solutions to HW#8 are now available in
the CMPSCI main office. Kazu has posted HW#7 solutions linked from
his page.
 Notes for the last lecture are now up.
 To answer a common question as HW#8 comes due, yes, it is
true that only the best seven of your eight homeworks count. So if
you are not likely to do as well as the worst of your first seven,
you have no incentive to hand in #8. Understanding the
solutions to
#8, though, will be important on the final.
 The practice final
is now up. The final exam will have the same format and cover
about the same material.
 The HW#8 assignment is now up.
I have extended the due date of HW#8 to Wednesday 14 May for oncampus
students. I will announce the due date for offcampus students after
consulting with Kazu, but it will be no earlier than Monday 19 May.
 Note that the web site contains postscript
versions of each lecture as well, named, e.g., "27.ps" rather than "27.pdf"
in the same directory.
 The
pointer on the syllabus to the original ChandraKozenStockmeyer
article on alternation, actually works and gets you a pdf file.
 I've posted the midterm exam
and solutions for the midterm in the
exam directory. The solutions include notes on some of the more common
wrong answers I received.
 Our final exam is on Monday 19 May from 8:0010:00 a.m. in
AEBN 119, as indicated on the University final exam schedule. I've
just fixed an incorrect date that has been on the
syllabus all term. (Apologies for the delay, especially
to the student who pointed out my mistake a month ago!)
 I've received and graded midterm exams from nine offcampus
students. The scores, in sorted order, were 39, 43, 44, 47, 51, 54, 63,
66, and 67. (See below for the scale and the distribution of oncampus
scores.) I will send back the exams by mail, but if you would like a
breakdown of your score and comments by email please let me know before
the start of next week.
 The HW#7 assignment is now up.

I have graded the 41 inclass midterms and will return them in class
tomorrow. Scores were widely distributed but I thought the test did
a good job of showing what you could and couldn't do. The median grade
was a 68, an AB. There were two 99's and a 98, and a 26. More
specifically:
 9099 (high A ): 5
 8089 (low A ): 7
 7079 (high AB): 8
 6069 (low AB): 9
 5059 (high B ): 5
 4049 (low B ): 4
 3039 (high BC): 2
 2029 (low BC): 1
 The Question and Answer Page now has several
questions on the practice midterm and the material for the real midterm.
 Midterm exams for offcampus students will be given to the
PEEAS office midday on Monday, just after the offcampus students have
taken the exam. Offcampus students should arrange to take the exam on
or about 8 April 2003, before viewing the lecture of 3 April 2003.
 My solution to the practice midterm
is now posted.
 My solution to HW#3 is posted here
and Kazu's solutions to HW#1 and HW#2 are linked from
Kazu's page.
There will be no posted solutions for HW#4.
 TA Kazu Hirata has established
a web page
for courserelated material. Offcampus students may confirm receipt
of their homeworks there.
 Remember that the lecture notes are very similar to
last year's notes, which are available for the whole term
here.
 The question and answer page logs questions
of general interest that I receive, primarily about the current homework
assignment.
In Spring 2002 this course was taught by Prof. Immerman,
and I will be modeling this term's offering closely on his.
The web site for his offering is still available
here.
In particular, the two textbooks for the course are
the same. These are:
 Papadimitriou, Computational Complexity (AddisonWesley, 1994)
 Barwise and Etchemendy, Language, Proof, and Logic
(University of Chicago Press, 2002) (Same as earlier edition through
Seven Bridges Press, 1999 but the 1999 edition is out of print)
Both texts are now available for sale at
Jeffrey Amherst Bookshop
in downtown Amherst. The B&E book ("LPL") will be used
for roughly the middle third of the course, beginning with Lecture #10
on Mon 3 Mar.
Last modified 28 August 2003