Miscellaneous · Q2: is injective
Show that the function given by is injective.
Let with . Then , so
where
If , then . If , both squares are 0, so and then . Again .
So . Hence is injective.
Want the reason behind every step, the basics and the checks?
Injective means: equal outputs force equal inputs. So we assume and must reach . The difference of two cubes factorises: . A product is zero only if a factor is zero, and the second factor, after completing the square, is a sum of two squares, which is zero only when . Either way, . Unlike squaring, cubing keeps the sign, so and do not collide.
- Given
- for every real , with domain and codomain .
- To show
- is injective (one-one): for all .
- Plan
- (1) Assume for arbitrary reals. (2) Move everything to one side and factorise. (3) Use the zero-product rule: two cases. (4) Show the second case also gives , by completing the square. (5) Conclude, and compare with .
Basics you need
is injective (one-one) if implies for all . To prove it, start from equal outputs for arbitrary inputs and deduce the inputs are equal. Basics: one-one and onto
. Check by expanding: .
If for real numbers , then or (or both). This is how a factorisation turns one hard equation into two easier ones.
For every real , , with only when . So a sum of squares is 0 only when . Completing the square is the standard way to write an expression as such a sum.
When every possibility falls into one of a few cases, proving the conclusion in each case proves it outright. Basics: proving and disproving
Solution
Assume equal outputs
Let and be any real numbers with , that is,
We must show that this forces .
Injective is an "if … then" statement about every pair of inputs, so the proof must start from the "if" part, equal outputs, for inputs that could be any real numbers. Checking examples such as and would not cover the infinitely many other pairs.Definition of injective (Basic 1): .Bring everything to one side and factorise
Subtract from both sides and use the difference-of-cubes identity (Basic 2) with :
The factor is exactly what we want to show is zero. Factorising makes it appear explicitly, so the question becomes: can the other factor be zero instead? We cannot just take cube roots of both sides here without first knowing that a real number has only one real cube root, which is essentially the statement we are trying to prove., checked by expanding the right-hand side (Basic 2).Two cases from the zero-product rule
A product of two real numbers is zero only when one of them is zero (Basic 3). So
Case 1: . Then immediately.
Case 2: . This needs more work, done in the next step.
The two cases cover every possibility, so if each one leads to , the proof is complete (Basic 5). Case 1 is the outcome we want; Case 2 is the only way the proof could fail, so we must examine it carefully rather than ignore it.Zero-product rule (Basic 3): or .Case 2: complete the square
Treat as the start of a square in . Since ,
Both terms on the right are (a square, and times a square). A sum of two non-negative numbers is 0 only when both are 0 (Basic 4), so
So in Case 2, , and in particular .
An expression like has a cross term that can be negative, so it is not obvious that it is never negative. Completing the square removes the cross term and writes the expression as a sum of squares, whose sign is obvious. This shows the second factor is positive except when both numbers are 0.for every real , with equality only at (Basic 4). Completing the square: .Conclude, and compare with
In both cases . Since were arbitrary, for all real , and is injective.
Compare squaring. For the same method gives , and the second factor can be zero with : take . Indeed , so is not one-one on . Cubing keeps the sign: but .
The graph of rises all the way from left to right, so every horizontal line meets it at most once (in fact exactly once). That is the picture of "injective". The parabola , by contrast, is met twice by the line , at . Showing where the same method breaks for makes clear what the proof really used: the second factor is never zero unless both numbers are zero, while is zero for every pair of opposite numbers.One counterexample ( but ) disproves one-one for (Basic 5).
If , then . Either , or the bracket is 0, which forces . So in every case, and is injective on . (The answer key gives no answer for this "show that" question.)
Take : the second factor is , and the completed-square form gives , the same and positive. The site's verification script cubes 81 different fractions from to and finds 81 different cubes, checks that for every sample pair other than , and confirms both identities symbolically.
- Dividing by without checking it is non-zero. It is zero when ; that case must be handled (it gives anyway).
- Claiming "because squares are positive". The middle term can be negative; completing the square is what proves the claim.
- "Taking cube roots" with no justification. That every real number has exactly one real cube root is essentially what we are proving.
- Proving the converse. is true for every function and proves nothing about injectivity.
- For a polynomial , proving one-one usually means: factor as and show the "something" cannot vanish when .
- Completing the square is the tool for showing a quadratic expression is never negative.
- is also onto (every real is the cube of ), so it is a bijection; compare Miscellaneous Q1, another bijection proved by solving .
Every fact used, and where it comes from
| Fact | Used in | Source |
|---|---|---|
| Injective (one-one) | Steps 1, 5 | Definition 5, Section 1.3 (Lesson: types of functions) |
| Step 2 | Algebraic identity (Class IX); verified symbolically in the site's verification script | |
| Zero-product rule | Step 3 | Property of real numbers |
| ; squares | Step 4 | Completing the square; verified symbolically |
| is not one-one on | Step 5 | Counterexample (compare Example 11, Section 1.3) |
| Sample cubes distinct; graph coordinates | Check, Figure | Computed in the site's verification script |
Tell us if something on this page is wrong or unclear, or if you would like a fuller explanation, another method or a new feature. Send feedback on this problem →