Recall that being a function means that every -value corresponds to exactly one -value.
Consider the example for . It is a function and satisfies the above condition. In particular, corresponds to exactly one -value . Similarly, .
In trying to set up the inverse, however, we run into a problem. The value of would correspond to both and , leaving us with more than one possible output for the inverse relation. Thus the inverse is not a function, and we say that does not exist.
For to exist, we then need each -value in the range of to correspond to a unique -value in the domain. We call such functions one-to-one.
One-to-one functions
Given a function its inverse function exists if is one-to-one.
Horizontal line test
Determining whether a function is one-to-one (and hence whether it has an inverse) can be done with the horizontal line test.
A function is one-to-one if all horizontal lines cut the graph of at most once.
A function is not one-to-one if there is a horizontal line that cuts the graph of more than once.