Large deviations for high-dimensional random projections of _p^n-balls

Abstract: The paper provides a description of the large deviation behavior for the Euclidean norm of projections of

e l l_{p}^{n}

-balls to high-dimensional random subspaces. More precisely, for each integer

n g e q 1

, let

k_{n} i n 1, l d o t s, n - 1

,

E^{(n)}

be a uniform random

k_{n}

-dimensional subspace of

m a t h b b R^{n}

and

X^{(n)}

be a random point that is uniformly distributed in the

e l l_{p}^{n}

-ball of

m a t h b b R^{n}

for some

p i n [1, i n f t y]

. Then the Euclidean norms

| P_{E^{(n)}} X^{(n)} |_{2}

of the orthogonal projections are shown to satisfy a large deviation principle as the space dimension

n

tends to infinity. Its speed and rate function are identified, making thereby visible how they depend on

p

and the growth of the sequence of subspace dimensions

k_{n}

. As a key tool we prove a probabilistic representation of

| P_{E^{(n)}} X^{(n)} |_{2}

which allows us to separate the influence of the parameter

p

and the subspace dimension

k_{n}

.

Recommendations

Cites work

Cited in

(26)

This page was built for publication: Large deviations for high-dimensional random projections of \(\ell_p^n\)-balls