Fixed point method for autonomous on-line neural network training

Authors:
PaweŁ WawrzyńSki;Bartosz Papis
Affiliations:
Warsaw University of Technology, Poland;Warsaw University of Technology, Poland
Venue:
Neurocomputing
Year:
2011

Citing 13
Cited 1

Terminal attractors in neural networks

Neural Networks
Acceleration Techniques for the Backpropagation Algorithm

Proceedings of the EURASIP Workshop 1990 on Neural Networks
A Very Fast Learning Method for Neural Networks Based on Sensitivity Analysis

The Journal of Machine Learning Research
Neighborhood based modified backpropagation algorithm using adaptive learning parameters for training feedforward neural networks

Neurocomputing
Real-time reinforcement learning by sequential Actor-Critics and experience replay

Neural Networks
Recursive Adaptation of Stepsize Parameter for Non-stationary Environments

PRIMA '09 Proceedings of the 12th International Conference on Principles of Practice in Multi-Agent Systems
Global synchronization for discrete-time stochastic complex networks with randomly occurred nonlinearities and mixed time delays

IEEE Transactions on Neural Networks
Adapting bias by gradient descent: an incremental version of delta-bar-delta

AAAI'92 Proceedings of the tenth national conference on Artificial intelligence
An improved training algorithm for feedforward neural network learning based on terminal attractors

Journal of Global Optimization
A real-time learning algorithm for a multilayered neural networkbased on the extended Kalman filter

IEEE Transactions on Signal Processing
Improvement of the neighborhood based Levenberg-Marquardt algorithm by local adaptation of the learning coefficient

IEEE Transactions on Neural Networks
On Adaptive Learning Rate That Guarantees Convergence in Feedforward Networks

IEEE Transactions on Neural Networks
Training feedforward networks with the Marquardt algorithm

IEEE Transactions on Neural Networks

2013 Special Issue: Autonomous reinforcement learning with experience replay

Neural Networks

Quantified Score

Hi-index	0.01

Visualization

Abstract

This paper considers on-line training of feedforward neural networks. Training examples are only available through sampling from a certain, possibly infinite, distribution. In order to make the learning process autonomous, one can employ Extended Kalman Filter or stochastic steepest descent with adaptively adjusted step-sizes. Here the latter is considered. A scheme of determining step-sizes is introduced that satisfies the following requirements: (i) it does not need any auxiliary problem-dependent parameters, (ii) it does not assume any particular loss function that the training process is intended to minimize, (iii) it makes the learning process stable and efficient. An experimental study with several approximation problems is presented. Within this study the presented approach is compared with Extended Kalman Filter and LFI, with satisfactory results.