16 lines
820 B
16 lines
820 B
2 years ago
|
|
||
|
|
Exercise 1.46: Several of the numerical methods
|
||
|
|
described in this chapter are instances of an extremely general computational
|
||
|
|
strategy known as
|
||
|
|
iterative improvement. Iterative improvement says
|
||
|
|
that, to compute something, we start with an initial guess for the answer, test
|
||
|
|
if the guess is good enough, and otherwise improve the guess and continue the
|
||
|
|
process using the improved guess as the new guess. Write a procedure
|
||
|
|
iterative-improve that takes two procedures as arguments: a method for
|
||
|
|
telling whether a guess is good enough and a method for improving a guess.
|
||
|
|
Iterative-improve should return as its value a procedure that takes a
|
||
|
|
guess as argument and keeps improving the guess until it is good enough.
|
||
|
|
Rewrite the sqrt procedure of 1.1.7 and the
|
||
|
|
fixed-point procedure of 1.3.3 in terms of
|
||
|
|
iterative-improve.
|