We introduce a construction of a uniform measure over a functional class B r which is similar to a Besov class with smoothness index r. We then consider the problem of approximating B r using a manifold M n which consists of all linear manifolds spanned by n ridge functions, i.e.,It is proved that for some subset A/B r of probabilistic measure 1&$, for all f # A the degree of approximation of M n behaves asymptotically as 1Ân rÂ(d&1) . As a direct consequence the probabilistic (n, $ )-width for nonlinear approximation denoted as d n, $ (B r , +, M n ), where + is a uniform measure over B r , is similarly bounded. The lower bound holds also for the specific case of approximation using a manifold of one hidden layer neural networks with n hidden units.
Academic Press