Volker Tresp, Ralph Neuneier, Hans Georg Zimmermann  Home/Search
Context Related
| tresp.org/./papers...ips_ebd_proc.ps.gz Cached: PS.gz PS PDF This document uses CoBlitz to cache paper downloads. If your firewall is blocking outgoing connections to port 3125, you can use these links to download local copies.
Image Update HelpPS.gz PS PDF From: tresp.org/ (more) (Enter author homepages) |
Rate this article:      (best)Comment on this article |
Abstract: Optimal Brain Damage (OBD) is a method for reducing the number of weights in a neural network. OBD estimates the increase in cost function if weights are pruned and is a valid approximation if the learning algorithm has converged into a local minimum. On the other hand it is often desirable to terminate the learning process before a local minimum is reached (early stopping). In this paper we show that OBD estimates the increase in cost function incorrectly if the network is not in a local... (Update)
Context of citations to this paper: More
...when a local minimum is reached and the cost is locally quadratic in this minimum. Both hypothesis are barely met in practice. Tresp et al. 1997) propose two weight pruning techniques from the same family, coined EBD (Early Brain Damage) and EBS (Early Brain Surgeon) They...
Cited by: More
Multi-Agent Market Modeling Based On Neural Networks - Grothmann (Correct)
Extended Kalman Filter Based Pruning Algorithms And Several Aspects .. - Sum (1998) (Correct)
Universal Distribution of Saliencies for Pruning in.. - Gorodkin, Hansen.. (1997) (Correct)
Similar documents based on text: More All
0.6: Optimal Brain Damage - Le Cun, Denker, Solla (1990) (Correct)
0.5: Connection Pruning with Static and Adaptive Pruning Schedules - Prechelt (1996) (Correct)
0.5: Comparing Adaptive and Non-Adaptive Connection Pruning with Pure .. - Prechelt (1996) (Correct)
Related documents from co-citation: More All
4: Second order derivatives for network pruning: Optimal brain surgeon - Hassibi, Stork - 1993
3: Generalization by weight elimination with application to forecasting (context) - Weigend, Rumelhart et al. - 1991
3: Multilayer feedforward networks are universal approximations (context) - Hornik, Stinchcombe et al. - 1989
BibTeX entry: (Update)
V. Tresp, R. Neuneier and H. G. Zimmermann, "Early brain damage," in Advances in Neural Information Processing Systems 9 (Denver, 1996) eds. M. C. Mozer, M. I. Jordan and T. Petsche, pp. 669--675. http://citeseer.ist.psu.edu/tresp96early.html More
@inproceedings{ tresp97early,
author = "Volker Tresp and Ralph Neuneier and Hans Georg Zimmermann",
title = "Early Brain Damage",
booktitle = "Advances in Neural Information Processing Systems",
volume = "9",
publisher = "The {MIT} Press",
editor = "Michael C. Mozer and Michael I. Jordan and Thomas Petsche",
pages = "669",
year = "1997",
url = "citeseer.ist.psu.edu/tresp96early.html" }
Citations not processed or no citations identified.
Documents on the same site (http://tresp.org/): More
Fisher Scoring and a Mixture of Modes Approach for.. - Briegel, Tresp (1999) (Correct)
Training Neural Networks with Deficient Data - Tresp (1994) (Correct)
Efficient Methods for Dealing with Missing Data in.. - Tresp, Neumeier, al. (1995) (Correct)
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC