Consider all pairs of polynomials f ( x ) and g ( x ) such that ⎩ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎧ f ( g ( x ) ) = f ( x ) + g ( x ) , f ( 0 ) = 5 , f ( 1 ) = 7 , g ( 0 ) = 0 . What are the last three digits of the sum of all (distinct) possible values of f ( 2 0 ) ?
Details and assumptions
The empty sum (sum of no elements) is 0.
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
This solution is essentially the same as James L.'s, but written more compactly. This is generally considered better, as long as all the important details are made explicit. A side remark one should use different letters for the degrees of the polynomials, because, say, f usually still means f ( x ) .
Let d f and d g be the degrees of f ( x ) and g ( x ) , respectively, and let f and g be the leading coefficients of f ( x ) and g ( x ) , respectively.
Note that f ( x ) is not a constant, since f ( 0 ) and f ( 1 ) take on different values. If g ( x ) is constant, then let g ( x ) = c . Then, we have
f ( c ) ⟹ f ( x ) = f ( x ) + c = f ( c ) − c ,
but f ( c ) − c is constant since c is a fixed value, so hence we get f ( x ) is constant, a contradiction. Hence, we have that d f , d g ≥ 1 .
Note that the degree of f ( g ( x ) ) is d f d g , and the degree of f ( x ) + g ( x ) is at most max ( d f , d g ) (The reason why we have the "max" is because if d f = d g and f = − g , then we don't have equality). From f ( g ( x ) ) = f ( x ) + g ( x ) , we get that d f d g ≤ max ( d f , d g ) .
If d f ≥ d g , then we get that
d f d g ⟹ d g ≤ d f ≤ 1 ,
so d g = 1 since g ( x ) is not constant.
Similarly, if d g ≥ d f , we get that d f = 1 .
Hence, either f ( x ) or g ( x ) is linear.
If f ( x ) is linear, then from f ( 0 ) = 5 and f ( 1 ) = 7 we get that f ( x ) = 2 x + 5 . Plugging this into f ( g ( x ) ) = f ( x ) + g ( x ) , we get that
2 g ( x ) + 5 ⟹ g ( x ) = g ( x ) + 2 x + 5 = 2 x ,
but this means that g ( 0 ) = 0 , contradicting g ( 0 ) = 0 .
Thus, we have that g ( x ) is linear, and f ( x ) has degree at least 2 . Let g ( x ) = a x + b .
Then, we get that f ( a x + b ) = f ( x ) + a x + b . Now, we compare the leading coefficients of both sides of this equation.
The leading coefficient of the left side is a d f f , and the leading coefficient of the right side is f since f ( x ) has degree at least 2 , so a x + b doesn't affect the leading coefficient. Hence,
a d f f ⟹ a d f ⟹ a = f = 1 = ± 1 .
If a = − 1 , then we get that f ( − x + b ) = f ( x ) − x + b . Plugging in x = b gives f ( 0 ) = f ( b ) = 5 , so define p ( x ) = f ( x ) − 5 . Then, we get that p ( 0 ) = p ( b ) = 0 . Note that since g ( 0 ) = 0 , we get that b = 0 . Since the degree of p ( x ) is at least 2 , we have that p ( x ) = x ( x − b ) q ( x ) for some polynomial q ( x ) .
We have that
f ( − x + b ) ⟹ f ( − x + b ) − 5 ⟹ p ( − x + b ) ⟹ ( − x + b ) ( − x ) q ( − x + b ) ⟹ ( − x ) q ( − x + b ) ⟹ x ( q ( x ) − q ( − x + b ) ) = f ( x ) − x + b = ( f ( x ) − 5 ) − x + b = p ( x ) − x + b = x ( x − b ) q ( x ) + ( − x + b ) = − x q ( x ) + 1 = 1
If q ( x ) = q ( − x + b ) , then we get that 0 = 1 , a contradiction. Otherwise, we have that the left side has degree at least 1 while the right side is constant, a contradiction.
Hence, we can't have g ( x ) = − x + b , so we must have g ( x ) = x + b . Unfortunately, a similar solution doesn't work for this case.
We have that f ( x + b ) = f ( x ) + x + b , or f ( x + b ) − f ( x ) = x + b . But f ( x + b ) − f ( x ) is just a finite difference of f ( x ) , so hence it has degree d f − 1 (Also, after expanding f ( x + b ) − f ( x ) , it is easy to see that the coefficient of x d f is 0 and the coefficient of x d f − 1 is b f d f , which is nonzero since b , f , d f = 0 ).
Hence, the left side of f ( x + b ) − f ( x ) = x + b has degree d f − 1 and the right side has degree 1 , so hence we have d f = 2 and thus f ( x ) is quadratic.
Let f ( x ) = r x 2 + s x + t . Since f ( 0 ) = 5 we have f ( x ) = r x 2 + s x + 5 . Since f ( 1 ) = 7 we have that r + s = 2 and f ( x ) = r x 2 + ( 2 − r ) x + 5 . Plugging this into f ( x + b ) − f ( x ) = x + b gives
[ r ( x + b ) 2 + ( 2 − r ) ( x + b ) + 5 ] − [ r x 2 + ( 2 − r ) x + 5 ] ⟹ 2 r b x + r b 2 + ( 2 − r ) b = x + b = x + b ,
so equating coefficients gives us 2 r b = 1 ⟹ r b = 2 1 and
r b 2 + ( 2 − r ) b ⟹ r b + 2 − r ⟹ 2 1 + 2 − r ⟹ r = b = 1 = 1 = 2 3 ,
where we divided by b = 0 and substituted r b = 2 1 . Then, we get that f ( x ) = 2 3 x 2 + 2 1 x + 5 and g ( x ) = x + 3 1 . It is easy to check that these two polynomials satisfy our original conditions, so our only possibility for f ( 2 0 ) is 2 3 ⋅ 2 0 2 + 2 1 ⋅ 2 0 + 5 = 6 1 5 .
This is a very detailed solution. While it may appear long, it is just because all the details are explicitly written up. There is no really short solution to this problem.
Consider the following, where de g ( f ( x ) ) denotes the degree of f ( x ) . de g ( f ( g ( x ) ) ) = de g ( f ( x ) + g ( x ) ) de g ( f ( x ) ) ⋅ de g ( g ( x ) ) = max ( de g ( f ( x ) ) , de g ( g ( x ) ) ) Therefore, one of ( f ( x ) , g ( x ) ) is linear. First, assume f ( x ) is linear. Therefore, f ( x ) = 2 x + 5 Plugging into the first equation: 2 g ( x ) + 5 = 2 x + 5 + g ( x ) g ( x ) = 2 x However, since 2 ( 0 ) = 0 , this does not satisfy the fourth given condition. Therefore, de g ( f ( x ) ) ≥ 2 and de g ( g ( x ) ) = 1 . Let g ( x ) = a x + b , and b = 0 by the fourth given condition. f ( a x + b ) = f ( x ) + a x + b Considering the leading term of both sides, a = 1 . f ( x + b ) − f ( x ) = x + b This expression is an algebraic identity. Let f ( x ) = a 1 x n + a 2 x n − 1 ⋯ Now, the coefficient of the term with degree n − 1 is a 1 n b on the LHS and either 0 or 1 on the right side. Therefore, n = 2 , since a 1 , n , b = 0 . Now, let f ( x ) = a 1 x 2 + a 2 x + 5 . We can expand and equate coefficients, giving: a 1 b = 2 1 a 1 b + a 2 = 1 Therefore, a 2 = 2 1 . f ( x ) = a 1 x 2 + 2 1 x + 5 Since f ( 1 ) = 7 , we find that a 1 = 2 3 . We now know only one polynomial f ( x ) satisfies the conditions. The answer is f ( 2 0 ) = 6 1 5 .
There seems to be a little glitch here: why is a = 1 ? If the degree of f is even, then a = − 1 is also possible,
Log in to reply
You are correct. Let me add a little:
If a = − 1 , we can still show n = 2 , and then we equate coefficients to get: a 2 + a 1 b = 1 2 a 2 + 2 a 1 b = 1 This is a contradiction, so a = 1 , and we can proceed as above.
First, let the degree of f be i and the degree of g be j. We see that i*j=max(i,j).
Suppose j>=i. We see that i*j=j, so, assuming g is nonconstant, i=1. This implies f=2x+5. Therefore, 2g+5=2x+5+g, so g=2x, which is impossible. If j=0, then we have that i=0, contradiction, since f is nonconstant.
Therefore, i>j. Since the first condition gives us that i*j=i. Since i>0 (because f is nonconstant), we have that j=1. Now, consider the leading coefficient of f(g(x)). We know it must equal the leading coefficient of f(x), so we see that g(x)=x+y, for some nonzero constant y. Furthermore, (x+y)^i has positive coefficients of degree i-1, so it follows that i=2.
Therefore f(x)=ax^2+bx+5. Substituting into our equation, ax^2+2ayx+ay^2+bx+by+5=ax^2+bx+5+x+y, so 2ayx+by+ay^2=x+y. Matching coefficients, 2ay=1, so ay=1/2. Furthermore, ay^2+by=y. Since y is nonzero, ay+b=1 and b=1/2.
Now we have f(x)=ax^2+x/2+5. We see that a=3/2 by the third initial condition, so the answer is
3/2 400+1/2 20+5=615
Problem Loading...
Note Loading...
Set Loading...
Denote the first equation by ( 1 ) . Let m a x ( a , b ) denote the maximum value of a and b. Let the degrees of f ( x ) and g ( x ) be f and g respectively. Note that f > 0 since f ( 0 ) = f ( 1 ) . By ( 1 ) , f g = m a x ( f , g ) . We now have two cases.
i) g ≥ f
Then f g = g . Now we have two cases.
a) g = 0
Then g ( x ) = c for some c ∈ R . But, by ( 1 ) , f ( x ) = f ( c ) − c , which is a constant, which contradicts f > 0 .
b) f = 1
Then f ( 0 ) = 5 and f ( 1 ) = 7 implies f ( x ) = 2 x + 5 . Substituting this into ( 1 ) gives g ( x ) = 2 x , which contradicts g ( 0 ) = 0 .
ii) f > g Then, f g = f . Since f > 0 , g = 1 . Let g ( x ) = a x + b . Let f ( x ) = a n x n + a n − 1 x n − 1 + . . . + a 0 , a n = 0 . The leading coefficient of f ( g ( x ) ) is a n ⋅ a n and the leading coefficient of f ( x ) + g ( x ) is a n . Thus, a n = 1 . We again have two cases.
a) a = 1
Then, since g ( 0 ) = 0 , b = 0 . Equation ( 1 ) gives f ( x + b ) − f ( x ) = x + b . The degree of the LHS is n − 1 while the degree of the RHS is 1 , so n = 2 . Note that a 0 = f ( 0 ) = 5 and a 1 + a 2 = f ( 1 ) − a 0 = 2 . Substituting this into f ( x + b ) − f ( x ) = x + b and solving for a 1 , a 2 and b gives a 1 = 2 1 , a 2 = 2 3 , b = 3 1 . It remains to note that the set of polynomials f ( x ) = 2 3 x 2 + 2 1 x + 5 , g ( x ) = x + 3 1 satisfies the equations. This gives f ( 2 0 ) = 6 1 5 .
b) a = − 1 and n is even.
Again, f ( − x + b ) − f ( x ) = − x + b and the degree of LHS is still n − 1 since n is even. Thus, n = 2 using the same argument as the above. Using similar arguments, we can find the equations a 1 + a 2 = 2 , − 2 b a 2 − 2 a 1 = − 1 , b 2 a 2 + b a 1 = b . Simplifying the last two equations give b a 2 + a 1 = 2 1 and b a 2 + a 1 = 1 , which is a contradiction. Thus, f ( 2 0 ) = 6 1 5 is the only solution.