Real vs. Complex Spectral Bases for Neural Operators: The Role of Green's Function Alignment

• URL: http://arxiv.org/abs/2606.24851v3
• Date: Mon, 29 Jun 2026 23:01:00 GMT
• Title: Real vs. Complex Spectral Bases for Neural Operators: The Role of Green's Function Alignment
• Abstract summary: We introduce the Hartley Neural Operator (HNO), the exact real-valued mirror of FNO.<n>Because the real Hartley spectrum is not halved by conjugate symmetry, HNO retains twice as many frequency corners as FNO.<n>We find an elliptic-versus-time-dependent split that is monotone in operator phase content.
• Score: 1.037435491622876
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Fourier Neural Operators (FNO) learn solution operators of partial differential equations by parameterizing global convolutions in the complex Fourier domain. For real-valued PDE solutions, the complex FFT carries representational redundancy through conjugate symmetry. We introduce the Hartley Neural Operator (HNO), the exact real-valued mirror of FNO: it replaces the FFT with the purely real Discrete Hartley Transform and learns a single real multiplier per retained spectral mode, with no complex arithmetic. Because the real Hartley spectrum is not halved by conjugate symmetry, HNO retains twice as many frequency corners as FNO but one real weight where FNO carries a complex pair, so the two operators are iso-parametric at equal width and differ only in spectral basis. Our central thesis is that the best basis is a property of the operator. Self-adjoint elliptic operators (Poisson, biharmonic) have real, symmetric Green's functions that the real Hartley multiplier diagonalizes exactly, and HNO is favored there. Time-dependent operators carry phase, from oscillation in the wave equation to transport in advection, Burgers, and Navier-Stokes, which a real diagonal multiplier cannot represent, so FNO is favored there, and increasingly so with the operator's phase content, leaving the phaseless heat equation as the borderline case. Training both operators identically and benchmarking across PDE classes, initial-condition families, and boundary conditions, we find an elliptic-versus-time-dependent split that is monotone in operator phase content and matches the Green's-function theory we develop. Rather than a universal winner, our findings give a predictive rule: match the spectral basis to the symmetry of the solution operator.

Related papers

• Finite-range Lattice Momentum Operators for Quantum Field Theory [0.3437656066916039]
  We propose a Z-transform framework for the analysis and synthesis of finite range lattice momentum operators in quantum field theory.<n>We show that near $= $ only plane waves propagate coherently, and these exhibit group velocities far exceeding the speed of light.<n>No ghost wave packet solutions exist near $= $.
  arXiv  Detail & Related papers  (2026-08-18T03:38:47Z)
• Isospectrality and Operator Complexity [0.0]
  We study a pair of exactly solvable, isospectral fermion chains, one strongly interacting and one quadratic.<n>Despite their identical spectra, the two models realize distinct phases and sharply different notions of operator complexity.
  arXiv  Detail & Related papers  (2026-06-03T18:00:05Z)
• The Double Well Done Doubly-Well [0.17188280334580197]
  symmetric double-well potential is one of the simplest quantum-mechanical systems in which perturbative and non-perturbative physics are deeply entangled.<n>Resurgence ties the two features together, organizing the exact spectrum into a single tightly-constrained trans-series.<n>This paper gives a self-contained account of this trans-series from two complementary approaches.
  arXiv  Detail & Related papers  (2026-06-03T18:00:01Z)
• Trainable Quantum Spectral Models for Partial Differential Equations [0.0]
  We study trainable quantum spectral models (QSMs) for solving linear partial differential equations (PDEs)<n>Instead of learning solutions directly in physical space, QSMs learn the inverse differential operator in a spectral representation, embedding prior knowledge of the equation's natural basis.<n>Our results identify operator-aware spectral representations as a promising route toward trainable and physically grounded quantum methods for scientific computing.
  arXiv  Detail & Related papers  (2026-05-29T12:47:50Z)
• Bose-Einstein thermal operators for semidefinite optimization [10.718144055666487]
  We recast general SDPs as bosonic free-energy minimization problems at strictly positive temperature, regularized by the Bose-Einstein entropy.<n>We develop hybrid quantum-classical algorithms for the regularized SDP using only Hamiltonian simulation.
  arXiv  Detail & Related papers  (2026-05-26T16:09:55Z)
• Comparison of the standard and dressed-picture master equations for the quantum Rabi model in the ultrastrong coupling regime [78.55450146498553]
  This chapter investigates the effects of relaxation and dephasing on the quantum Rabi model in the ultrastrong coupling regime.<n>It provides explicit formulas to implement and numerically solve the resulting nonunitary dynamics.
  arXiv  Detail & Related papers  (2026-04-10T01:22:07Z)
• Parallel Complex Diffusion for Scalable Time Series Generation [50.01609741902786]
  PaCoDi is a spectral-native architecture that decouples generative modeling in the frequency domain.<n>We show that PaCoDi outperforms existing baselines in both generation quality and inference speed.
  arXiv  Detail & Related papers  (2026-02-10T14:31:53Z)
• SpectraKAN: Conditioning Spectral Operators [21.190440188964452]
  We introduce SpectraKAN, a neural operator that conditions the spectral operator on the input itself, turning spectral into an input-conditioned integral operator.<n>This is achieved by extracting a compact global representation from static-temporal history and using it to modulate a multi-scale trunk via single-query cross-attention.<n>Across diverse PDE benchmarks, SpectraKAN achieves state-of-the-art performance, reducing RMSE by up to 49% over strong baselines, with particularly large gains on challenging-temporal prediction tasks.
  arXiv  Detail & Related papers  (2026-02-05T01:30:25Z)
• SFO: Learning PDE Operators via Spectral Filtering [25.390393080966422]
  We introduce a neural operator that parameterizes integral kernels using the Universal Spectral Basis (USB)<n>By learning only the spectral coefficients of rapidly decaying eigenvalues, SFO achieves a highly efficient representation.
  arXiv  Detail & Related papers  (2026-01-23T10:45:52Z)
• Quantum simulation of the Fokker-Planck equation via Schrodingerization [33.76659022113328]
  This paper studies a quantum simulation technique for solving the Fokker-Planck equation. We employ the Schrodingerization method-it converts any linear partial and ordinary differential equation with non-Hermitian dynamics into systems of Schrodinger-type equations.
  arXiv  Detail & Related papers  (2024-04-21T08:53:27Z)
• Group Equivariant Fourier Neural Operators for Partial Differential Equations [33.71890280061319]
  We consider solving partial differential equations with Fourier neural operators (FNOs) In this work, we extend group convolutions to the frequency domain and design Fourier layers that are equivariant to rotations, translations, and reflections. The resulting $G$-FNO architecture generalizes well across input resolutions and performs well in settings with varying levels of symmetry.
  arXiv  Detail & Related papers  (2023-06-09T06:34:16Z)
• Factorized Fourier Neural Operators [77.47313102926017]
  The Factorized Fourier Neural Operator (F-FNO) is a learning-based method for simulating partial differential equations. We show that our model maintains an error rate of 2% while still running an order of magnitude faster than a numerical solver.
  arXiv  Detail & Related papers  (2021-11-27T03:34:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.