Robust control methods for on-line statistical learning.
Summary: The issue of controlling that data processing in an experiment results not affected by the presence of outliers is relevant for statistical control and learning studies. Learning schemes should thus be tested for their capacity of handling outliers in the observed training set so to achieve reliable estimates with respect to the crucial bias and variance aspects. We describe possible ways of endowing neural networks with statistically robust properties by defining feasible error criteria. It is convenient to cast neural nets in state space representations and apply both Kalman filter and stochastic approximation procedures in order to suggest statistically robustified solutions for on-line learning.
- Neural networks and statistical inference: seeking robust and efficient learning
- Robust learning in a partial least-squares neural network
- A robust on‐line learning algorithm for intelligent control systems
- scientific article; zbMATH DE number 1784956 (Why is no real title available?)
- Pointwise probability reinforcements for robust statistical inference
- On‐line learning neural networks for sensor validation for the flight control system of a B777 research scale model
- Dynamical systems approach to outlier robust deep neural networks for regression
- Advances in Neural Networks – ISNN 2005
- Robust sequential learning of feedforward neural networks in the presence of heavy-tailed noise
This page was built for publication: Robust control methods for on-line statistical learning.
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1347548)
