The DOI matches its Crossref record; 13 of the 14 references it cites were found in a registry or an index.
Overfitting and neural networks: conjugate gradient and backpropagation
Abstract
Methods for controlling the bias/variance tradeoff typically assume that overfitting or overtraining is a global phenomenon. For multi-layer perceptron (MLP) neural networks, global parameters such as the training time, network size, or the amount of weight decay are commonly used to control the bias/variance tradeoff. However, the degree of overfitting can vary significantly throughout the input space of the model. We show that overselection of the degrees of freedom for an MLP trained with backpropagation can improve the approximation in regions of underfitting, while not significantly overfitting in other regions. This can be a significant advantage over other models. Furthermore, we show that "better" learning algorithms such as conjugate gradient can in fact lead to worse generalization, because they can be more prone to creating varying degrees of overfitting in different regions of the input space. While experimental results can not cover all practical situations, our results do help to explain common behavior that does not agree with theoretical expectations. Our results suggest one important reason for the relative success of MLPs, bring into question common beliefs about neural network training regarding training algorithms, overfitting, and optimal network size, suggest alternate guidelines for practical use, and help to direct future work.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“This framework has proven its capability in overcoming some of the main challenges of object detection, including overtraining problems [5, 6] and false alarm distribution caused by the patches’ proposal stage.”
From A survey on deep multimodal learning for computer vision: advances, trends, applications, and datasets (2022) · cites this paper as [5]
Citations
Cites
Advances in neural information processing systems 7
(1996) · Computers & Mathematics with Applications · doi:10.1016/s0898-1221(96)90463-0 · Cited by 14,759
Neural Networks and the Bias/Variance Dilemma
Geman, S., Bienenstock, E., & Doursat, R. (1992) · Neural Computation · doi:10.1162/neco.1992.4.1.1 · Cited by 3,578
(1989) · Cited by 3,214
Advances in Neural Information Processing Systems 37
Touretzky, D. S. (2024) · doi:10.52202/079017 · Cited by 2,874
Advances in Neural Information Processing Systems 5
Zemel, R. S., Williams, C. K. I., & Mozer, M. C. (1993) · Cited by 1,896
What Size Net Gives Valid Generalization?
Baum, E. B., & Haussler, D. H. (1989) · Neural Computation · doi:10.1162/neco.1989.1.1.151 · Cited by 1,549
(1992) · Cited by 1,405
Moody, J. E. (1991) · Cited by 437
For Valid Generalization the Size of the Weights is More Important than the Size of the Network
Bartlett, P. L. (1996) · Cited by 213
Note on Learning Rate Schedules for Stochastic Optimization
Darken, C. J., & Moody, J. E. (1990) · neural information processing systems · Cited by 133
Research works printed but not linked: 4
- darken, 1991, Advances in neural information processing systems, 3, 832
- weigend, 1991, Advances in neural information processing systems, 3, 875
- le cun, 1990, Advances in neural information processing systems, 2, 598
- vapnik, 1982, Estimation of Dependencies Based on Empirical Data
Cited by
Sorted by citations
Prediction of cardiovascular risk factors from retinal fundus photographs via deep learning
Poplin, R., Varadarajan, A. V., Blumer, K., Liu, Y., McConnell, M. V., Corrado, G. S., Peng, L., & Webster, D. R. (2018) · Nature Biomedical Engineering · doi:10.1038/s41551-018-0195-0 · Cited by 1,865
An analysis of particle swarm optimizers
van den Bergh, F., & Engelbrecht, A. P. (2002) · UpSpace Institutional Repository (University of Pretoria) · Cited by 1,411
Deep learning classifiers for hyperspectral imaging: A review
Paoletti, M. E., Haut, J. M., Plaza, J., & Plaza, A. (2019) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2019.09.006 · Cited by 1,144
A deep learning system for detecting diabetic retinopathy across the disease spectrum
Dai, L., Wu, L., Li, H., Cai, C., Wu, Q., Kong, H., Liu, R., Wang, X., Hou, X., Liu, Y., et al. (2021) · Nature Communications · doi:10.1038/s41467-021-23458-5 · Cited by 620
A systematic review on overfitting control in shallow and deep neural networks
Bejani, M. M., & Ghatee, M. (2021) · Artificial Intelligence Review · doi:10.1007/s10462-021-09975-1 · Cited by 624
Metadata
- Authors
- Lawrence, S., & Giles, C. L.
- Affiliations
- NEC (Japan) (JP)
- ORCID
- Giles, C.L.
- Venue
- Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks. IJCNN 2000. Neural Computing: New Challenges and Perspectives for the New Millennium
- Publisher
- IEEE
- Published
- 2000
- Pages
- 114-119 vol.1
- ISSN
- 2161-4407 (ISSN-L)
- OpenAlex
- W2114367267
- Cited by
- 491
- References
- 14
- Citations by year
- 2026: 11 · 2025: 32 · 2024: 41 · 2023: 42 · 2022: 51 · 2021: 67 · 2020: 60 · 2019: 44 · 2018: 23 · 2017: 14 · 2016: 6 · 2015: 8 · 2014: 10 · 2013: 10 · 2012: 8
- FWCI
- 2.54
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- International Joint Conference on Neural Networks (IJCNN) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Stochastic Gradient Optimization Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and ELM
- Keywords
- overfitting, bias-variance trade-off, multilayer perceptron, conjugate gradient method, backpropagation, generalization, learning algorithms, network size, weight decay, input space
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
