Title :
On stochastic approximation algorithms for classes of PAC learning problems
Author :
Rao, Nageswara S V ; Uppuluri, V.R.R. ; Oblow, E.M.
Author_Institution :
Center for Eng. Syst. Adv. Res., Oak Ridge Nat. Lab., TN, USA
fDate :
6/1/1997 12:00:00 AM
Abstract :
The classical stochastic approximation methods are shown to yield algorithms to solve several formulations of the PAC learning problem defined on the domain [0,1]d. Under some smoothness conditions on the probability measure functions, simple algorithms to solve some PAC learning problems are proposed based on networks of nonpolynomial units (e.g. artificial neural networks). Conditions on the sizes of the samples required to ensure the error bounds are derived using martingale inequalities
Keywords :
inference mechanisms; learning (artificial intelligence); neural nets; PAC learning problems; artificial neural networks; error bounds; martingale inequalities; nonpolynomial units; probability measure functions; smoothness conditions; stochastic approximation algorithms; Approximation algorithms; Approximation methods; Artificial neural networks; Backpropagation algorithms; Concrete; Neural networks; Pattern recognition; Power engineering and energy; Risk management; Stochastic processes;
Journal_Title :
Systems, Man, and Cybernetics, Part B: Cybernetics, IEEE Transactions on
DOI :
10.1109/3477.584958
