Pith

open record

sign in

arxiv: 2006.10739 · v1 · pith:6QWJAYDE · submitted 2020-06-18 · cs.CV · cs.LG

Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 reserved pith:6QWJAYDErecord.jsonopen to challenge →

classification cs.CV cs.LG
keywords fourierlearncomputerdomainsfeaturefeaturesfunctionsgraphics
0
0 comments X
read the original abstract

We show that passing input points through a simple Fourier feature mapping enables a multilayer perceptron (MLP) to learn high-frequency functions in low-dimensional problem domains. These results shed light on recent advances in computer vision and graphics that achieve state-of-the-art results by using MLPs to represent complex 3D objects and scenes. Using tools from the neural tangent kernel (NTK) literature, we show that a standard MLP fails to learn high frequencies both in theory and in practice. To overcome this spectral bias, we use a Fourier feature mapping to transform the effective NTK into a stationary kernel with a tunable bandwidth. We suggest an approach for selecting problem-specific Fourier features that greatly improves the performance of MLPs for low-dimensional regression tasks relevant to the computer vision and graphics communities.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 26 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Sampling the Schwinger Model with Gauge-Equivariant Diffusion

    hep-lat 2026-06 unverdicted novelty 7.0

    A gauge-equivariant diffusion model samples Schwinger model configurations, yielding unbiased observables matching MCMC and qualitatively less topological freezing than HMC.

  2. One Generator, Any Process: LLM-Conditioning for the LHC

    hep-ph 2026-06 unverdicted novelty 7.0

    LLM embeddings condition generative networks for LHC events, yielding faster convergence, higher quality, and generalization to unseen processes.

  3. Physics informed operator learning of parameter dependent spectra

    gr-qc 2026-04 unverdicted novelty 7.0

    DeepOPiraKAN learns parameter-to-spectrum mappings via operator learning and achieves relative errors of O(10^{-6}) to O(10^{-4}) for Kerr black hole quasinormal modes up to n=7 when benchmarked against Leaver's method.

  4. Deep learning-based phase-field modelling of brittle fracture in anisotropic media

    physics.comp-ph 2026-03 unverdicted novelty 7.0

    A variational physics-informed neural network solves higher-order anisotropic phase-field fracture models by minimizing total energy with B-spline enriched trial functions.

  5. NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction

    cs.CV 2021-06 unverdicted novelty 7.0

    NeuS introduces a bias-free volume rendering method for signed distance function representations to reconstruct accurate surfaces from 2D images.

  6. Self-Supervised Implicit CEST Reconstruction via Physics-Informed Lorentz Encoding

    cs.LG 2026-07 conditional novelty 6.0

    Embedding learnable Lorentzian basis functions into the spectral encoding layer of an implicit neural representation enables high-fidelity CEST Z-spectrum reconstruction from as few as 21 frequency offsets.

  7. Confidence-feedback-weighted graph matching network: online-offline laser-induced damage site matching under complex interference

    cs.CV 2026-06 unverdicted novelty 6.0

    A confidence-feedback-weighted graph matching network achieves 96.36% F1-score on damage site matching by using matchability confidence to weight edge features and applying geometric consistency and hard-example mining.

  8. One Generator, Any Process: LLM-Conditioning for the LHC

    hep-ph 2026-06 unverdicted novelty 6.0

    LLM embeddings condition a generative transformer to enable faster convergence, better performance, and generalization to unseen LHC processes using a single model.

  9. Multi-Resolution Tactile Imitation Learning for Contact-Rich Robotic Manipulation

    cs.RO 2026-06 unverdicted novelty 6.0

    MiTaS fuses multi-resolution tactile data from GelSight and Evetac sensors with vision using modality-specific stems and transformer fusion to condition flow-matching policies, reporting 80% average success on five co...

  10. IV-Net: A neural network for elliptic PDEs with random and highly varying coefficients

    math.NA 2026-05 unverdicted novelty 6.0

    IV-Net is a multigrid-inspired convolutional neural operator that approximates solutions to linear elliptic PDEs with high-contrast coefficients and shows better accuracy than POD and other neural operators on heterog...

  11. RoPeSLR: 3D RoPE-driven Sparse-LowRank Attention for Efficient Diffusion Transformers

    cs.CV 2026-05 unverdicted novelty 6.0

    RoPeSLR combines 3D RoPE-guided sparse attention with head-wise low-rank parameterization to achieve sub-quadratic complexity in DiTs while preserving distance awareness for efficient ultra-long video synthesis.

  12. LiFT: Lifted Inter-slice Feature Trajectories for 3D Image Generation from 2D Generators

    cs.CV 2026-05 unverdicted novelty 6.0

    LiFT factorizes 3D medical volume synthesis into per-slice 2D generation and inter-slice trajectory learning, using a tri-planar drifting loss for unconditional coherence and a z-context mixer for paired translation tasks.

  13. MoMo: Conditioned Contrastive Representation Learning for Preference-Modulated Planning

    cs.LG 2026-05 unverdicted novelty 6.0

    MoMo conditions contrastive representations and prediction operators on user preferences via FiLM and low-rank modulation to enable continuous modulation of plan safety while preserving inference efficiency.

  14. MoMo: Conditioned Contrastive Representation Learning for Preference-Modulated Planning

    cs.LG 2026-05 unverdicted novelty 6.0

    MoMo uses Feature-Wise Linear Modulation and low-rank neural modulation to condition contrastive planning representations on user preferences while preserving inference efficiency and probability density ratios.

  15. From Characterization To Construction: Generative Quantum Circuit Synthesis from Gate Set Tomography Data

    quant-ph 2026-05 unverdicted novelty 6.0

    A generative QMLC framework tokenizes GST data, embeds it via curriculum-trained set-vision transformers into a context-aware latent space, and uses diffusion models to synthesize circuits conditioned on desired measu...

  16. When PINNs Go Wrong: Pseudo-Time Stepping Against Spurious Solutions

    cs.LG 2026-04 conditional novelty 6.0

    PINNs fail on spurious solutions admitted by the residual loss; adaptive pseudo-time stepping with Jacobian-based step selection improves accuracy and robustness on PDE benchmarks.

  17. A Deep Ritz Method for High-Dimensional Steady States of the Cahn-Hilliard Equation

    math.NA 2026-04 unverdicted novelty 6.0

    A Deep Ritz method with augmented Lagrangian and Fourier feature mappings computes high-dimensional steady states of the Cahn-Hilliard equation and identifies multiple nontrivial phase separation patterns.

  18. AE-ViT: Stable Long-Horizon Parametric Partial Differential Equations Modeling

    cs.LG 2026-04 unverdicted novelty 6.0

    AE-ViT combines a convolutional autoencoder with a latent-space transformer and multi-stage parameter plus coordinate injection to deliver stable long-horizon predictions for parametric PDEs, cutting relative rollout ...

  19. Monte Carlo Event Generation with Continuous Normalizing Flows

    hep-ph 2026-04 conditional novelty 6.0

    Continuous normalizing flows improve unweighting efficiency in Monte Carlo event generation for high-jet-multiplicity collider processes by factors up to 184, with wall-time gains of about ten when combined with coupl...

  20. Cell-induced densification and tether formation in fibrous extracellular matrices with biomimetic physics-informed neural networks

    cs.LG 2026-03 unverdicted novelty 6.0

    Bio-PINNs with a near-to-far curriculum and deformation-uncertainty proxy recover cell-induced densified phases and tether morphologies more reliably than standard adaptive PINN baselines in single-cell and multicellu...

  21. Nomic Embed: Training a Reproducible Long Context Text Embedder

    cs.CL 2024-02 conditional novelty 6.0

    Nomic AI produced and open-sourced a reproducible 8192-context English text embedder that exceeds OpenAI Ada-002 and text-embedding-3-small performance on MTEB short-context and LoCo long-context benchmarks.

  22. General-Purpose Nonlinear Function Approximation via Linear Integrated Photonics

    cs.ET 2026-06 unverdicted novelty 5.0

    Optical random Fourier feature mapping enables approximation of tenth-order polynomials, special functions, neural activations, and high-dimensional softmax in linear silicon photonic hardware.

  23. FEDONet : Fourier-Embedded DeepONet for Spectrally Accurate Operator Learning

    cs.LG 2025-09 conditional novelty 5.0

    FEDONet augments DeepONet with Fourier-embedded trunk networks using random Fourier features, yielding lower L2 reconstruction errors than standard DeepONet on Burgers', 2D Poisson, Eikonal, Allen-Cahn, and Kuramoto-S...

  24. Flow Matching for Convective-Scale Precipitation Downscaling

    physics.ao-ph 2026-05 unverdicted novelty 4.0

    Flow matching produces better spatial structure than diffusion models for convective precipitation downscaling but underestimates heavy rainfall amounts.

  25. Machine Learning-based Separation of the He I 10830{\AA} Chromospheric Signal: Quantitative Analysis of Chromosphere-Corona Intensity in the Quiet Sun

    astro-ph.SR 2026-05 unverdicted novelty 4.0

    CNN separation of He I 10830Å chromospheric signal from photospheric contamination in quiet Sun reveals R ≈ -0.84 anti-correlation with 304Å and magnetic-field-dependent EUV coupling.

  26. A Deep Ritz Method for High-Dimensional Steady States of the Cahn-Hilliard Equation

    math.NA 2026-04 conditional novelty 4.0

    A deep learning framework computes multiple steady states of the Cahn-Hilliard equation in high dimensions using Fourier feature mappings and an augmented Lagrangian for mass conservation.