SCXT 350

 

 

 

 

Hour Exam #1

 

 

 

 

 

Name_______________________________________

 

 

 

 

 

Friday, Oct. 4

100 pts.

 

 

 


The story so far:

 

1.         (10 pts.) One thread of the story begins with Descartes.  He is fascinated with automata, but is unable to come up with an explanation of human behavor in terms of automata.  What is the stumbling block?  What does he say about it?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2.         (15 pts.)          This thread of the story continues with the 1950 paper by Alan Turing, in which he describes the Turing Test.  Give a brief description of the test, including a brief diagram.

 


3.         (10 pts.)          Turing is really making an argument for machine intelligence.  He provides up with several "objections" which he then refutes.  List and describe two of them, with (briefly) Turing's response.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The second thread of the story has to do with an assumption made by Turing: that intelligent action is computable.  We need to explore what this assumption means.  What does it mean to say that something is computable?  In particular,

 

4.      (10 pts.)          What is an algorithm?

 


5.      (15 pts.)          A computer is sometimes defined as a machine that runs algorithms.  In our first look at computers, we looked at a simple diagram for a computer consisting of an arithmetic logic unit, a control unit, memory, something for input and output, and something for long term storage.  Briefly reproduce that sketch (5 pts. out of the 15) and give brief definitions of the arithmetic logic unit, memory, and the control unit.

 


6.      (10 pts.)          Computers need to be instructed on what to do (programmed).  We have taken a brief look at the programming language Lisp.  In that language, how would we write expressions

 

         a.         To calculate

 

 

 

 

 

 

         b.         To find the third item in a list lst (which has at least three elements) using only CAR and CDR.

 

 

 

 

 

 

 


7.      (10 pts.)          Ok, so we may be (appropriately) skeptical about a full-scale commitment to the idea that intelligent behavior is computable.  But it might make sense to go part-way and think of using an information processing system as a model for some parts of human cognition.  Marr suggests that we consider three "levels of explanation" for an information processing system.  What are they, and what effect does one level have on the others?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

9.      (10 pts.)          We interact with information processing systems all the time.  Briefly, how would you apply Marr's analysis to an ATM?