The DOI matches its Crossref record; 37 of the 37 references it cites were found in a registry or an index.
Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks
Open accessAbstract
We study the training process of Deep Neural Networks (DNNs) from the Fourier analysis perspective. We demonstrate a very universal Frequency Principle (F-Principle) — DNNs often fit target functions from low to high frequencies — on high-dimensional benchmark datasets such as MNIST/CIFAR10 and deep neural networks such as VGG16. This F-Principle of DNNs is opposite to the behavior of Jacobi method, a conventional iterative numerical scheme, which exhibits faster convergence for higher frequencies for various scientific computing problems. With theories under an idealized setting, we illustrate that this F-Principle results from the smoothness/regularity of the commonly used activation functions. The F-Principle implies an implicit bias that DNNs tend to fit training data by a low-frequency function. This understanding provides an explanation of good generalization of DNNs on most real datasets and bad generalization of DNNs on parity function or a randomized dataset.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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
Very Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
Partial Differential Equations
(1988) · Lecture Notes in Mathematics · doi:10.1007/bfb0082920 · Cited by 23,383
An overview of statistical learning theory
Vapnik, V. N. (1999) · IEEE Transactions on Neural Networks · doi:10.1109/72.788640 · Cited by 6,266
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 3×
“In the literature, on the one hand, there are different approaches attempting to mainly explain why the training process often leads to a DNN of good generalization ability even when the number of parameters is much larger than the number of training data (Zhang et al., 2016).”
BackgroundRademacher and Gaussian Complexities: Risk Bounds and Structural Results
Bartlett, P. L., & Mendelson, S. (2001) · Lecture Notes in Computer Science · doi:10.1007/3-540-44581-1_15 · Cited by 1,949
The Deep Ritz Method: A Deep Learning-Based Numerical Algorithm for Solving Variational Problems
E, W., & Yu, B. (2018) · Communications in Mathematics and Statistics · doi:10.1007/s40304-018-0127-z · Cited by 1,418
A Closer Look at Memorization in Deep Networks
Arpit, D., Jastrzębski, S., Ballas, N., Krueger, D., Bengio, E., Kanwal, M. S., Maharaj, T., Fischer, A., Courville, A., Bengio, Y., et al. (2017) · arXiv · doi:10.48550/arxiv.1706.05394 · Cited by 1,307
E, W., Han, J., & Jentzen, A. (2017) · Communications in Mathematics and Statistics · doi:10.1007/s40304-017-0117-6 · Cited by 822
On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima
Keskar, N. S., Mudigere, D., Nocedal, J., Smelyanskiy, M., & Tang, P. T. P. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1609.04836 · Cited by 525
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
DeepXDE: A Deep Learning Library for Solving Differential Equations
Lu, L., Meng, X., Mao, Z., & Karniadakis, G. E. (2021) · SIAM Review · doi:10.1137/19m1274067 · Cited by 2,512
When and why PINNs fail to train: A neural tangent kernel perspective
Wang, S., Yu, X., & Perdikaris, P. (2022) · Journal of Computational Physics · doi:10.1016/j.jcp.2021.110768 · Cited by 1,496
Meng, X., & Karniadakis, G. E. (2020) · Journal of Computational Physics · doi:10.1016/j.jcp.2019.109020 · Cited by 713
Focal Frequency Loss for Image Reconstruction and Synthesis
Jiang, L., Dai, B., Wu, W., & Loy, C. C. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01366 · Cited by 459
Metadata
- Authors
- John Xu, Z.-Q., Zhang, Y., Luo, T., Xiao, Y., & Ma, Z.
- Venue
- Communications in Computational Physics
- Publisher
- Global Science Press
- Published
- 2020-06
- Volume
- 28
- Issue
- 5
- Pages
- 1746-1767
- ISSN
- 1815-2406 (ISSN-L)
- 1991-7120
- OpenAlex
- W2914483840
- Cited by
- 522
- References
- 37
- FWCI
- 17.71
- 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: Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks. https://app.cochat.ai/papers/10.4208/cicp.oa-2020-0085
- The paper
- John Xu, Z.-Q., Zhang, Y., Luo, T., Xiao, Y., & Ma, Z. (2020). Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks. Communications in Computational Physics. https://doi.org/10.4208/cicp.oa-2020-0085
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.