Law of the iterated logarithm for a constant-gain linear stochastic gradient algorithm

SIAM Journal on Control and Optimization

Volumn 39, Issue 2, 2000, Pages 533-570

  Joslin, J A ^a   Heunis, A J ^a  

^a Siemens Canada Limited   (Canada)

Author keywords

Functional law of the iterated logarithm; L mixing processes; Stochastic gradient algorithm

Indexed keywords

ALGORITHMS; FUNCTION EVALUATION; GAIN CONTROL; RANDOM PROCESSES; STOCHASTIC CONTROL SYSTEMS;

ITERATED LOGARITHMS; STOCHASTIC GRADIENT ALGORITHMS;

GRADIENT METHODS;

EID: 0034453883     PISSN: 03630129     EISSN: None     Source Type: Journal    
DOI: 10.1137/S0363012997331007     Document Type: Article

References (27)

1. Weak convergence and local stability properties of fixed step size recursive algorithms
2. On Strassen's version of the log log law
3. Limit theorems for sums of weakly dependent Banach space valued random variables
4. On a class of mixing processes
5. Multiple integrals with respect to L-mixing processes
6. Asymptotic properties of prediction error estimators in approximate system identification
7. Rates of convergence for an adaptive filtering algorithm driven by stationary dependent data
8. On stochastic processes defined by differential equations with a small parameter
9. Rates of convergence in a central limit theorem for stochastic processes defined by differential equations with a small parameter
10. A law of the iterated logarithm for stochastic processes defined by differential equations with a small parameter
11. A strong convergence theorem for Banach space valued random variables
12. Weak convergence and asymptotic properties of adaptive filters with constant gains
13. Convergence analysis of parametric identification methods
14. Asymptotic normality of prediction error estimators for approximate system models
15. General moment and probability inequalities for the maximum partial sum
16. Moment inequalities and the strong law of large numbers
17. Strong diffusion approximations for recursive stochastic algorithms
18. An invariance principle for the law of the iterated logarithm
19. On the error of the Gaussian approximation for convolutions