Disclaimer: I'm a beginner with inverse functions.
Can anyone explain what I'm doing wrong here? I'd like to avoid using "y" -- that is, I want to show everything in terms of x and f(x).
Thanks!
Answer
This is a very good question!
One of the main viewpoints, when it comes to inverse functions, is that they make "input" and "output" switch places - if (x,f(x)) is a point on the graph of y=f(x), then the point (f(x),x) is a point on the graph of y=f−1(x).
Algebraically, we can think of this as replacing x (our old input) with f−1(x) (our new output), and f(x) (our old output) with x (our new input): We start thinking of f(x) as our "input", and x as the "output", f−1(x). What makes things confusing is that we agree to always use "x" for our input, which means x gets a little overused here.
Since that reads clunkier than I'd intended, how about a table?
InputOutputOldxf(x)Newf(x)xNew Renamedxf−1(x)
Going from Old to New, we're making x and f(x) switch places (since we start thinking about the inverse function). Then we rename things by calling our input x and our output f−1(x). You could even think of this renaming in an algebraic way, by applying f−1 to what we have:
f(x)↦f−1(f(x))=xx↦f−1(x).
So, when you reach the step with x=1f(x), you've already effectively switched input and output (by solving for the old output, f(x), which becomes the new input). Now you just need to rename:
x=1f(x)rename⟶f−1(x)=1x.
As you can see, the study of functions requires new ways of thinking; things are more abstract. While in the end we can reduce this process to a sequence steps to be followed, it's more subtle than, for example, the "just multiply both sides by 2" sort of algebra you have more practice with.
No comments:
Post a Comment