[Home]Function/Suggestion

HomePage | Function | Recent Changes | Preferences

Difference (from prior major revision) (no other diffs)

Changed: 1,79c1

Mathematical function




A mathematical function (also called mapping) is a means of
associating every object in a certain set X with a unique object in
a certain set Y. The set X is called the domain of the function,
and the set Y is called the codomain of the function. The subset
of Y that consists of all elements in Y that are associated with
some elements in X is called the range of the function. The unique element
in Y that is associated by a function f with x is usually
written as f(x).

In Set Theory a function from a set X to a set Y is defined as a
binary relation over X and Y that is both functional (i.e.
associates with every element in X at most one element in Y) and
total (i.e. associates with every element in X at least one element
of Y). If the binary relation is only functional then it is called a
partial function.

Functions can be specified in basically three ways. The simplest way is to
simply enumerate the associated pairs. (Which is only possible if the domain
is finite.) For example the function Weight that assigns to every living
human being in the United States their weight in pounds, might be enumerated
as follows.
* Larry -> 160, or Weight(Larry) = 160
* Jimmy -> 165, or Weight(Jimmy) = 165
* Ruth -> 125, or Weight(Ruth) = 125
* Cindy -> 120, or Weight(Cindy) = 120
* . . .
This sometimes also called specification by extension.

The second way of specifying a function is by giving an algorithm or an
[algebraic expresssion]? that specifies how for every element in the domain
the element in the codomain is determined. For example the following defines
a function f:

: f(x) = x5 - 1

(For a calculus of functions that are specified by algorithms
see Lambda calculus.)

The third way of specifying a function is by giving an equation that
relates variables that are associated with the domain and the codomain. The
function f above, for example, can be specified with the following
equation.

: y = x5 - 1

The variable that is associated with the domain (here x) is called the independent variable and the variable associated with the codomain (here y) is called
the dependent variable.

The last two ways of specifying a function are sometimes also called
specification by intension.

Note that although originally the notion of function was
primarily used to associate numbers with numbers, it can be used to
associate any mathematical construct with any other mathematical construct
such as vectors, matrices and all kinds of sets. An common example of this
are vector valued functions that always result in a vector.

There are 4 basic kinds of functions. Let f be a function from X to
Y then f is:
# an into function if for some y in Y there is no x in X such that f(x) = y,
# onto function or surjection if for every y in Y there is an x in X such that f(x) = y,
# one-to-one function or injection if for every y in Y there is one and only one x in X such that f(x) = y, and
# bijection if it is both "onto" and "one-to-one," or is both a surjection and an injection.

Examples:

* The function abs from Z (the set of Integers) to Z defined by

:: abs(x) = | x |

: where | x | is the absolute value of x, is "into." Since there exists an integer -5 in Z such that there is no integer i in Z such that abs(i) = -5, abs is "into."
* Consider the function abs* from Z to Z* (the set of non-negative integers) that is defined by the same rule as abs. Then abs* is "onto," or a surjection, since for every y in Z*, that is, every non-negative integer, there is at least one x in Z such that abs*(x) = y.
* Consider the function add1 from Z to Z that is specified by the rule
:: add1(x) = x + 1

: then add1 is "one-to-one" or an injection, since for every y in Z there is one and only one x in Z such that add1(x) = y.
* Consider the function add1, as defined above. Then add1 is both "onto" and "one-to-one" or a bijection, since it is already "onto" and we can show it is "one-to-one" as well. Thus, given a y in Z, there is only one x in Z such that add1(x) = y, that is, x = y - 1.



HomePage | Function | Recent Changes | Preferences
This page is read-only | View other revisions
Last edited July 23, 2001 6:10 am by Jan Hidders (diff)
Search: