Computer Science 320: Problem Solving
Spring 1995
Course description and Syllabus

Instructor: Bruce J. MacLennan
Phone: 4-5067
Office: 110-A, Ayres Hall
Email: maclennan@cs.utk.edu
Hours: 2:30-3:25 or make an appointment.

Teaching Assistant: Chuck Hand
Phone: 4-8635
Office: 1B, Ayres Hall
Email: hand@cs.utk.edu
Hours: 10:00-12:00 in Hydra Lab

Classes: 3:35-4:25 WF (Ayres 316 - change from scheduled!)
Labs: 2:30-5:30 M (Ayres 15)

CS 320 Home Page

Information

Prerequisites - important!
Texts
Assignments
Laboratory
Grading
Policy on joint work and outside help
Goals
General Approach
Caveats
Topics
Schedule
MacLennan's home page
Computer Science at UTK
UTK Home Page

Prerequisites

CS 112 (structured programming and data structures; familiarity with Unix and X-windows)
Math 221 and CS 311 (mathematics of computer science)
Math 222 (probability theory)

If you lack any of these prerequisites, you must see me to discuss whether or not you should take the course.

Required Texts

Algorithmics by Giles Brassard and Paul Bratley. This will be the source of regular reading assignments and frequent homework exercises. This text is very challenging: it could easily be used in a graduate course. However, it reads freshly and is chock full of good ideas. It should be enjoyable and worthwhile reading, but expect the lectures and labs to bring the material back down to earth. Used copies should be readily available. This book has also been used in the 400-level CS course on algorithms.
Common LISPcraft by Robert Wilensky. This will be your primary source for learning the language in which your programs will be written. It will also be the source of occasional homework exercises. You should bring this book to laboratory each week. Used copies should be readily available.

Optional Text

How To Read and Do Proofs (Second Edition) by Daniel Solow. This will be the source of optional reading assignments and homework exercises. We will briefly cover all the book. This book has also been used in CS 311.

Assignments

Readings as assigned each week with the homework.

Mandatory attendance at laboratory (see below).
Weekly homework exercises, handed out at lab and due at the beginning of the lab the following week. There are no homeworks in weeks in which a test occurs. Most of the homeworks will be paper-and-pencil work, but some will include programming (in Lisp) and other work on the machine. I will hand out answers to each homework soon after it is due, so late homeworks will not normally be accepted.
Two tests (1 hour, closed book).
A final examination.

Laboratory

Attendance at laboratory is mandatory. You may make up one missed lab, simply by making arrangements with the instructor. If you miss a lab and do not make arrangements to make it up, you will fail the course.

We attempt to design labs so that a well-prepared student who works energetically can complete the lab in the 3-hour lab session. You should strive to do so, since the teaching assistant will be at the lab to answer your questions. However, you may continue to work on the lab after the lab session; details will be given at the first lab.

Grading Plan

Lab: 20% Weekly homework: 10% First test: 20% Second test: 20% Final exam: 30%

Doing the weekly homework is necessary to succeed in this course. In a sense, it counts far more than merely 10%. Likewise, succeeding at the labs is necessary to succeed in the tests.

Policy on joint work and outside help

The ability to share common interests and ideas is a valuable tool for learning. I encourage you to discuss your homework and lab problems with your fellow students, and to seek outside help when you do not understand something, be it Lisp code or algorithm analysis. However, I insist you abide by two rules. First, you must always make a serious attempt to understand the problem and a reasonable attempt to solve it by yourself before you ask for help or discuss the problem with someone else. Second, whatever you turn in, you must completely understand. It is acceptable, for instance, for someone else to explain to you how to get over some programming hurdle, but you must understand the answer given (and not just notice that it does the job) if you turn in that work.

If you collaborate with someone on an assignment (per the above rules), and if the collaboration is substantial, please note on the assignment that I worked with so-and-so. This will serve as a gentle reminder of the above rules.

Goals

Become reasonably adept at programming in Lisp. [practical]
Learn standard ``methods'' for inventing algorithms and become skilled at applying such methods. [practical / theoretical]
Learn the fundamental language and skills of algorithm analysis. [theoretical / practical]
Learn how to read and write proofs. [theoretical]
Receive an inkling of the ``great ideas'' in Computer Science, enough to claim to ``know something'' about the science of computers. [philosophical / intellectual]

Each of the above goals is of roughly equal importance. However, the proficiency expected is in the order listed: respectable at Lisp, modest at algorithm invention and analysis, passable at proofs, and an intuitive feel for the ``great ideas.''

General Approach

Typically, the first lecture will introduce a topic in the abstract, followed quickly by several examples. In the laboratory, students will study the ``nuts and bolts'' of the topic, often by coding up an algorithm and running an experiment with it. Treatment of topics will be light. The emphasis will be on getting major ideas across, and letting other courses build upon a firm foundation created in CS 320.

Caveats

This course is likely to take more time and effort than most of your other courses have.
CS 112, CS 311 and Math 221-222 are prerequisites for this course. If you had trouble in those courses, you will have to work doubly hard in this course.

Topics

Introduction to problem solving by human and computer (Brassard & Bratley, ch. 1)
Proof techniques (Solow)
Analyzing the efficiency of algorithms (Brassard & Bratley, ch. 2)
Greedy algorithms (ch. 3)
Divide-and-conquer algorithms (ch. 4)
Dynamic programming (ch. 5)
Probabilistic algorithms (ch. 8)
Genetic algorithms
Neural networks
Introduction to decidability and complexity (ch. 10)