From Tarski's plank problem to simultaneous approximation

Abstract: A {em slab} (or plank) of width

w

is a part of the

d

-dimensional space that lies between two parallel hyperplanes at distance

w

from each other. It is conjectured that any slabs

S_{1}, S_{2}, l d o t s

whose total width is divergent have suitable translates that altogether cover

m a t h b b R^{d}

. We show that this statement is true if the widths of the slabs,

w_{1}, w_{2}, l d o t s

, satisfy the slightly stronger condition

l i m s u p_{n i g h t a r r o w i n f t y} f r a c w_{1} + w_{2} + l d o t s + w_{n} l o g (1 / w_{n}) > 0

. This can be regarded as a converse of Bang's theorem, better known as Tarski's plank problem. We apply our results to a problem on simultaneous approximation of polynomials. Given a positive integer

d

, we say that a sequence of positive numbers

x_{1} l e x_{2} l e l d o t s

{em controls} all polynomials of degree at most

d

if there exist

y_{1}, y_{2}, l d o t s i n m a t h b b R

such that for every polynomial

p

of degree at most

d

, there exists an index

i

with

| p (x_{i}) - y_{i} | l e q 1.

We prove that a sequence has this property if and only if

s u m_{i = 1}^{i n f t y} f r a c 1 x_{i}^{d}

is divergent. This settles an old conjecture of Makai and Pach.

Recommendations

Cites work

Cited in

(5)

This page was built for publication: From Tarski's plank problem to simultaneous approximation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4575386)