Scaling description of generalization with number of parameters in deep learning
Open accessThe DOI matches its Crossref record; 11 of the 11 references it cites were found in a registry or an index.
Abstract
Supervised deep learning involves the training of neural networks with a large number N of parameters. For large enough N , in the so-called over-parametrized regime, one can essentially fit the training data points. Sparsity-based arguments would suggest that the generalization error increases as N grows past a certain threshold N * . Instead, empirical studies have shown that in the over-parametrized regime, generalization error keeps decreasing with N . We resolve this paradox through a new framework. We rely on the so-called Neural Tangent Kernel, which connects large neural nets to kernel methods, to show that the initialization causes finite-size random fluctuations of the neural net output function f N around its expectation . These affect the generalization error for classification: under natural assumptions, it decays to a plateau value in a power-law fashion ∼ N −1/2 . This description breaks down at a so-called jamming transition N = N * . At this threshold, we argue that diverges. This result leads to a plausible explanation for the cusp in test error known to occur at N * . Our results are confirmed by extensive empirical observations on the MNIST and CIFAR image datasets. Our analysis finally suggests that, given a computational envelope, the smallest generalization error is obtained using several networks of intermediate sizes, just beyond N * , and averaging their outputs.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Hinton, G., Deng, L., Yu, D., Dahl, G., Mohamed, A.-R., Jaitly, N., Senior, A., Vanhoucke, V., Nguyen, P., Sainath, T., et al. (2012) · IEEE Signal Processing Magazine · doi:10.1109/msp.2012.2205597 · Cited by 10,260
Understanding deep learning requires rethinking generalization
Zhang, C., Bengio, S., Hardt, M., Recht, B., & Vinyals, O. (2016) · arXiv · doi:10.48550/arxiv.1611.03530 · Cited by 2,838 · cited 1×
“Surprisingly from the point of view of traditional statistical learning theory [5] DNNs generalize very well in practice, even without an explicit regularization.”
BackgroundReconciling modern machine-learning practice and the classical bias–variance trade-off
Belkin, M., Hsu, D., Ma, S., & Mandal, S. (2019) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1903070116 · Cited by 1,595
Neural Tangent Kernel: Convergence and Generalization in Neural Networks
Jacot, A., Gabriel, F., & Hongler, C. (2018) · arXiv · doi:10.48550/arxiv.1806.07572 · Cited by 1,562
Self-citationIdentifying and attacking the saddle point problem in high-dimensional non-convex optimization
Dauphin, Y., Pascanu, R., Gulcehre, C., Cho, K., Ganguli, S., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1406.2572 · Cited by 827 · cited 1×
“Numerical and theoretical studies [6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17] show that in the overparametrized regime, the loss landscape of DNNs is not rough with isolated minima as initially thought [18, 19], but instead has connected level sets and presents many flat directions, even near its global minimum.”
BackgroundThe Loss Surfaces of Multilayer Networks
Choromanska, A., Henaff, M., Mathieu, M., Arous, G. B., & LeCun, Y. (2014) · arXiv · doi:10.48550/arxiv.1412.0233 · Cited by 733
The generalization error of random features regression: Precise asymptotics and double descent curve
Mei, S., & Montanari, A. (2019) · arXiv · doi:10.48550/arxiv.1908.05355 · Cited by 537
Cited by
Physics-informed machine learning
Karniadakis, G. E., Kevrekidis, I. G., Lu, L., Perdikaris, P., Wang, S., & Yang, L. (2021) · Nature Reviews Physics · doi:10.1038/s42254-021-00314-5 · Cited by 8,718
Scaling Laws for Neural Language Models
Kaplan, J., McCandlish, S., Henighan, T., Brown, T. B., Chess, B., Child, R., Gray, S., Radford, A., Wu, J., & Amodei, D. (2020) · arXiv · doi:10.48550/arxiv.2001.08361 · Cited by 9,550
“We train language models on WebText2, an extended version of the WebText [RWC + 19] dataset, tokenized using byte-pair encoding [SHB15] with a vocabulary size n vocab = 50257 .”
BackgroundMethodInfluentialDeep double descent: where bigger models and more data hurt*
Nakkiran, P., Kaplun, G., Bansal, Y., Yang, T., Barak, B., & Sutskever, I. (2021) · Journal of Statistical Mechanics: Theory and Experiment · doi:10.1088/1742-5468/ac3a74 · Cited by 565
“Moreover, Geiger et al. (2019a) provide preliminary results for model-wise double descent in convolutional networks trained on CIFAR-10.”
BackgroundMethodResultInfluentialThe generalization error of random features regression: Precise asymptotics and double descent curve
Mei, S., & Montanari, A. (2019) · arXiv · doi:10.48550/arxiv.1908.05355 · Cited by 537
Surprises in high-dimensional ridgeless least squares interpolation
Hastie, T., Montanari, A., Rosset, S., & Tibshirani, R. J. (2022) · The Annals of Statistics · doi:10.1214/21-aos2133 · Cited by 503
Metadata
- Authors
- Geiger, M., Jacot, A., Spigler, S., Gabriel, F., Sagun, L., d’Ascoli, S., Biroli, G., Hongler, C., & Wyart, M.
- Venue
- Journal of Statistical Mechanics: Theory and Experiment
- Publisher
- IOP Publishing
- Published
- 2020-02-01
- Volume
- 2020
- Issue
- 2
- Pages
- 023401
- ISSN
- 1742-5468 (ISSN-L)
- OpenAlex
- W2907047316
- Cited by
- 155
- References
- 11
- FWCI
- 10.89
- Type
- journal article
- Language
- en
- Open access
- green
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Scaling description of generalization with number of parameters in deep learning. https://app.cochat.ai/papers/10.1088/1742-5468/ab633c
- The paper
- Geiger, M., Jacot, A., Spigler, S., Gabriel, F., Sagun, L., d’Ascoli, S., Biroli, G., Hongler, C., & Wyart, M. (2020). Scaling description of generalization with number of parameters in deep learning. Journal of Statistical Mechanics: Theory and Experiment. https://doi.org/10.1088/1742-5468/ab633c
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.