A
Andreas S. Weigend
Researcher at Stanford University
Publications - 10
Citations - 2342
Andreas S. Weigend is an academic researcher from Stanford University. The author has contributed to research in topics: Network complexity & Time series. The author has an hindex of 8, co-authored 10 publications receiving 2307 citations. Previous affiliations of Andreas S. Weigend include PARC.
Papers
More filters
Journal ArticleDOI
Predicting the future: a connectionist approach
TL;DR: Since the ultimate goal is accuracy in the prediction, it is found that sigmoid networks trained with the weight-elimination algorithm outperform traditional nonlinear statistical approaches.
Proceedings Article
Generalization by Weight-Elimination with Application to Forecasting
TL;DR: This work adds a term to the back propagation cost function that penalizes network complexity, called weight-elimination, and uses this procedure to predict the sunspot time series and the notoriously noisy series of currency exchange rates.
Journal Article
Bayesian Back-Propagation.
Wray Buntine,Andreas S. Weigend +1 more
TL;DR: In this article, approximate Bayesian methodems to statistical components of back-propagation-based networks are presented, which can be used both for nonlinear regression and for discrete one-of-C classification.
Book ChapterDOI
Back-propagation, weight-elimination and time series prediction
Andreas S. Weigend,Andreas S. Weigend,David E. Rumelhart,David E. Rumelhart,Bernardo A. Huberman,Bernardo A. Huberman +5 more
TL;DR: This work analyzes the sunspot series as an example of a real world time series of limited record length and finds that sigmoid networks trained with weight-elimination outperform traditional nonlinear statistical approaches.