Asymptotic properties of stochastic approximation procedures in the case of correlated noise (Q1082766)
From MaRDI portal
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Asymptotic properties of stochastic approximation procedures in the case of correlated noise |
scientific article |
Statements
Asymptotic properties of stochastic approximation procedures in the case of correlated noise (English)
0 references
1984
0 references
The asymptotic properties of a scalar stochastic approximation procedure are investigated in the case of correlated noise that contaminates the observations of the gradient of the function that is being optimized. Sufficient conditions of convergence with probability 1 and in the mean- square are obtained, as well as conditions that ensure asymptotic normality. The convergence rate is calculated, and examples of various noise correlation are discussed.
0 references
mean square convergence
0 references
scalar stochastic approximation
0 references
correlated noise
0 references
Sufficient conditions
0 references
convergence with probability 1
0 references
asymptotic normality
0 references
convergence rate
0 references
