V
Vitaly Maiorov
Researcher at Technion – Israel Institute of Technology
Publications - 27
Citations - 881
Vitaly Maiorov is an academic researcher from Technion – Israel Institute of Technology. The author has contributed to research in topics: Sobolev space & Upper and lower bounds. The author has an hindex of 14, co-authored 27 publications receiving 788 citations.
Papers
More filters
Journal ArticleDOI
Almost Linear VC Dimension Bounds for Piecewise Polynomial Networks
TL;DR: Upper and lower bounds on the VC dimension and pseudodimension of feedforward neural networks composed of piecewise polynomial activation functions are computed and it is shown that if the number of layers is fixed, then theVC dimension and pseudo-dimension grow as W log W, where W is thenumber of parameters in the network.
Journal ArticleDOI
Lower bounds for approximation by MLP neural networks
Vitaly Maiorov,Allan Pinkus +1 more
TL;DR: It is proved that there exists an analytic, strictly monotone, sigmoidal activation function for which this lower bound is essentially attained and that one can approximate arbitrarily well any continuous function on any compact domain by a two hidden layer MLP using a fixed finite number of units in each layer.
Journal ArticleDOI
On Best Approximation by Ridge Functions
TL;DR: In this paper, the authors considered the best approximation of some function classes by the manifold Mn consisting of sums of n arbitrary ridge functions and proved that the deviation of the Sobolev class, d2 from the manifold in the space L2 behaves asymptotically as n?r/(d?1).
Journal ArticleDOI
Approximation by neural networks and learning theory
TL;DR: It is shown that the least-squares estimator is almost-optimal for the problem of Learning Neural Networks from samples and can be used to solve Smale's network problem.
Journal ArticleDOI
On the Degree of Approximation by Manifolds of Finite Pseudo-Dimension
Vitaly Maiorov,Joel Ratsaby +1 more
TL;DR: Tight upper and lower bounds on ρn (Wr,dp, Lq) are obtained, which proves that approximation of Wr,dp by the family of manifolds of pseudo-dimension n is as powerful as approximation by theFamily of all nonlinear manifolds with continuous selection operators.