Proving Limit of Bilinear Function to 0

  • Thread starter Thread starter yifli
  • Start date Start date
  • Tags Tags
    Function
Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
3 replies · 2K views
yifli
Messages
68
Reaction score
0
Prove that [tex]\mbox{lim}_{(h,k) \rightarrow 0} \frac{|f(h,k)|}{|(h,k)|} = 0[/tex], where f is a bilinear function from R^n X R^m -> R^p

I don't know where to start only given the information that f is bilinear.
 
Physics news on Phys.org
micromass said:
Hi yifli! :smile:

Start by putting

[tex]h=h_1e_1+...+h_ne_n[/tex]

and

[tex]k=k_1e_1+...+k_me_m[/tex]

where the ei form a basis.

Now, how can you write f(h,k) now?

so f(h,k) can be written as follows:
[tex]f(h,k)=h_1(k_1f(e_1,\overline{e_1})+\cdots+k_mf(e_1,\overline{e_m}))+h_2(k_1f(e_2,\overline{e_1})+ \cdots +k_mf(e_2,\overline{e_m}))+\cdots+h_n(k_1f(e_n,\overline{e_1})+\cdots+k_mf(e_n,\overline{e_m}))[/tex]

e_i and \bar{e_i} represents the basis of R^n and R^m respectively

when both hi and ki tend to zero, I guess the numerator gets closer to zero faster than denominator because of the product hiki.
but how do you argue this rigorously?
 
yifli said:
so f(h,k) can be written as follows:
[tex]f(h,k)=h_1(k_1f(e_1,\overline{e_1})+\cdots+k_mf(e_1,\overline{e_m}))+h_2(k_1f(e_2,\overline{e_1})+ \cdots +k_mf(e_2,\overline{e_m}))+\cdots+h_n(k_1f(e_n,\overline{e_1})+\cdots+k_mf(e_n,\overline{e_m}))[/tex]

e_i and \bar{e_i} represents the basis of R^n and R^m respectively

when both hi and ki tend to zero, I guess the numerator gets closer to zero faster than denominator because of the product hiki.
but how do you argue this rigorously?

Indeed, let me illustrate this on an example. I hope the general method will be clear from that:

Take

[tex]f:\mathbb{R}\times\mathbb{R}\rightarrow \mathbb{R}:(x,y)\rightarrow xy[/tex]

Then we need to calculate

[tex]\lim_{(h,k)\rightarrow 0}{\frac{|hk|}{|(h,k)|}}[/tex]

The easiest thing to do is pick the norm (it works without this too).

[tex]|(h,k)|=\max\{h,k\}[/tex]

Then we know that [itex]h\leq \max\{h,k\}[/itex], thus

[tex]\frac{|h|}{|(h,k)|}\leq 1[/tex]

Hence

[tex]0\leq \lim_{(h,k)\rightarrow 0}{\frac{|hk|}{|(h,k)|}}\leq \lim_{(h,k)\rightarrow 0}{|k|}=0[/tex]

Try to do this in the general case...