Find all r>0 such that whenever f:R2→R
is a differentiable function such that
∣gradf(0,0)∣=1 and
∣gradf(u)−gradf(v)∣≤∣u−v∣
for all u,v∈R2, then the maximum of f on the disk
{u∈R2:∣u∣≤r} is attained at exactly one
point. (gradf(u)=(∂1f(u),∂2f(u)) is the gradient vector of f at the point u. For a vector
u=(a,b), ∣u∣=a2+b2.)
Solution (official)
To get an upper bound for r, set
f(x,y)=x−2x2+2y2. This function
satisfies the conditions, since
gradf(x,y)=(1−x,y),
gradf(0,0)=(1,0) and
∣gradf(x1,y1)−gradf(x2,y2)∣=∣(x2−x1,y1−y2)∣=∣(x1,y1)−(x2,y2)∣.
In the disk Dr={(x,y):x2+y2≤r2}f(x,y)=2x2+y2−(x−21)2+41≤2r2+41.
If r>21 then the absolute maximum is
2r2+41, attained at the points
(21,±r2−41). Therefore,
it is necessary that r≤21 because if
r>21 then the maximum is attained twice.
Suppose now that r≤1/2 and that f attains its maximum on
Dr at u,v, u=v. Since
∣gradf(z)−gradf(0)∣≤r,
∣gradf(z)∣≥1−r>0 for all z∈Dr. Hence
f may attain its maximum only at the boundary of Dr, so we must
have ∣u∣=∣v∣=r and gradf(u)=au and
gradf(v)=bv, where a,b≥0. Since
au=gradf(u) and bv=gradf(v)
belong to the disk D with centre gradf(0) and
radius r, they do not belong to the interior of Dr. Hence
∣gradf(u)−gradf(v)∣=∣au−bv∣≥∣u−v∣ and this inequality is strict since
D∩Dr contains no more than one point. But this contradicts
the assumption that
∣gradf(u)−gradf(v)∣≤∣u−v∣. So all r≤21 satisfies
the condition.
How the field did
contestants scored
226
average (of 20)
1.11
solved (≥ 80%)
1.8%
near-0 (≤ 10%)
91.6%
discrimination
0.30
Score distribution (field cohort)
Computed on contestants with a meaningful total (field cohort); discrimination is the corrected item–total correlation.