Abstract | ||
---|---|---|
Let φ be a univariate 2π-periodic function. Suppose that s ≥ 1 and f is a 2π-periodic function of s real variables. We study sufficient conditions in order that a neural network having a single hidden layer consisting of n neurons, each with an activation function φ, can be constructed so as to give a mean square approximation to f within a given accuracy ∈n, independent of the number of variables. We also discuss the case in which the activation function φ is not 2π-periodic. |
Year | DOI | Venue |
---|---|---|
1994 | 10.1147/rd.383.0277 | IBM Journal of Research and Development |
Keywords | Field | DocType |
neural network | Mean square,Discrete mathematics,Computer science,Activation function,Stochastic neural network,Electronic engineering,Univariate,Artificial neural network,Periodic graph (geometry) | Journal |
Volume | Issue | ISSN |
38 | 3 | 0018-8646 |
Citations | PageRank | References |
32 | 5.33 | 3 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Mhaskar, H.N. | 1 | 32 | 5.33 |
charles a micchelli | 2 | 47 | 7.73 |