The Mechanics of CNN Filtering with Rectification

• URL: http://arxiv.org/abs/2512.24338v1
• Date: Tue, 30 Dec 2025 16:44:41 GMT
• Title: The Mechanics of CNN Filtering with Rectification
• Abstract summary: This paper proposes elementary information mechanics as a new model for understanding the mechanical properties of convolutional filtering with rectification.<n>Even components cause image content to diffuse isotropically while preserving the center of mass, analogously to rest or potential energy with zero net momentum.<n>Odd kernels cause directional displacement of the center of mass, analogously to kinetic energy with non-zero momentum.
• Score: 0.6587738982080448
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: This paper proposes elementary information mechanics as a new model for understanding the mechanical properties of convolutional filtering with rectification, inspired by physical theories of special relativity and quantum mechanics. We consider kernels decomposed into orthogonal even and odd components. Even components cause image content to diffuse isotropically while preserving the center of mass, analogously to rest or potential energy with zero net momentum. Odd kernels cause directional displacement of the center of mass, analogously to kinetic energy with non-zero momentum. The speed of information displacement is linearly related to the ratio of odd vs total kernel energy. Even-Odd properties are analyzed in the spectral domain via the discrete cosine transform (DCT), where the structure of small convolutional filters (e.g. $3 \times 3$ pixels) is dominated by low-frequency bases, specifically the DC $Σ$ and gradient components $\nabla$, which define the fundamental modes of information propagation. To our knowledge, this is the first work demonstrating the link between information processing in generic CNNs and the energy-momentum relation, a cornerstone of modern relativistic physics.

Related papers

• A Unified Framework for the Mechanics of Information in Convolutional Neural Network Image Space [0.8632722291552638]
  This paper introduces a unified mathematical framework for modeling information propagation through convolutional neural networks (CNNs)<n>A correspondence is presented linking discrete filter symmetry and the relativistic energy--momentum relation under the widely used nonlinear rectified convolution operation.
  arXiv  Detail & Related papers  (2026-08-26T19:46:00Z)
• Learning the Kohn-Sham map with neural operators for quasi-linear scaling density functional theory [57.90994363143731]
  Kohn--Sham density functional theory (DFT) underpins electronic-structure simulations.<n>Repeated orbital diagonalizations lead to cubic scaling, restricting quantum calculations to modest scales only.<n>We develop a learning target for orbital-free DFT that maps a Kohn--Sham potential directly to the corresponding density and noninteracting kinetic energy.
  arXiv  Detail & Related papers  (2026-08-24T23:05:42Z)
• WorldParticle: Unified World Simulation of Lagrangian Particle Dynamics via Transformer [107.36949516442782]
  We present a learning-based particle simulator built on a single transformer architecture to model cloth, elastic solds, Newtonian and non-Newtonian fluids, granular materials, and molecular dynamics.
  arXiv  Detail & Related papers  (2026-05-14T18:18:12Z)
• Quantization Mapping on Dirac Dynamics via Voltage-Driven Charge Density in Monolayer Graphene: A Klein Paradox and Entropy-Ruled Wavevector Mechanics Study [0.0]
  We map the energy quantization for Dirac materials through the empirical relation of voltage-driven charge density in monolayer graphene.<n>This study reveals information about the interaction potential-DOS relationship in the Dirac materials.
  arXiv  Detail & Related papers  (2026-02-12T05:49:29Z)
• Dynamic transition of the generalized Jaynes-Cummings model: multi-particles and inter-particle interaction effects [1.305563895311544]
  How environments affect dynamics of quantum systems remains a central question.<n>We study the dynamic transition for the case that $N$ particles are initially excited with the boson modes in their vacuum state.
  arXiv  Detail & Related papers  (2024-12-30T01:47:51Z)
• Dynamics of Correlations and Entanglement Generation in Electron-Molecule Inelastic Scattering [0.0]
  The dynamics and processes involved in particle-molecule scattering, including nuclear dynamics, are described and analyzed.<n>The main process studied is the interatomic coulombic electronic capture (ICEC), an inelastic process that can lead to dissociation of the target molecule.
  arXiv  Detail & Related papers  (2024-11-15T17:08:05Z)
• Spin migration in density functional theory: energy, potential and density perspectives [0.0]
  We study the behavior of the energy, Kohn-Sham orbitals, the KS potentials and the electron density, with respect to fractional spin, in different atomic systems. Our results are instrumental for the assessment of the quality of existing approximations from a new perspective.
  arXiv  Detail & Related papers  (2024-11-03T12:49:35Z)
• Quantum tomography of helicity states for general scattering processes [55.2480439325792]
  Quantum tomography has become an indispensable tool in order to compute the density matrix $rho$ of quantum systems in Physics. We present the theoretical framework for reconstructing the helicity quantum initial state of a general scattering process.
  arXiv  Detail & Related papers  (2023-10-16T21:23:42Z)
• Molecular Geometry-aware Transformer for accurate 3D Atomic System modeling [51.83761266429285]
  We propose a novel Transformer architecture that takes nodes (atoms) and edges (bonds and nonbonding atom pairs) as inputs and models the interactions among them. Moleformer achieves state-of-the-art on the initial state to relaxed energy prediction of OC20 and is very competitive in QM9 on predicting quantum chemical properties.
  arXiv  Detail & Related papers  (2023-02-02T03:49:57Z)
• Spin Current Density Functional Theory of the Quantum Spin-Hall Phase [59.50307752165016]
  We apply the spin current density functional theory to the quantum spin-Hall phase. We show that the explicit account of spin currents in the electron-electron potential of the SCDFT is key to the appearance of a Dirac cone.
  arXiv  Detail & Related papers  (2022-08-29T20:46:26Z)
• Many Body Density of States of a system of non interacting spinless fermions [0.0]
  We introduce a new approach to evaluate the many-body density of states (MBDoS) in the case of systems that can be mapped into free fermions. We show that the many body spectrum can be expanded as a weighted sum of spectra given by the principal components of the filling matrix. We illustrate our method in two classes of problems that are mapped into spinless fermions.
  arXiv  Detail & Related papers  (2022-07-30T20:47:50Z)
• Transformer with Implicit Edges for Particle-based Physics Simulation [135.77656965678196]
  Transformer with Implicit Edges (TIE) captures the rich semantics of particle interactions in an edge-free manner. We evaluate our model on diverse domains of varying complexity and materials.
  arXiv  Detail & Related papers  (2022-07-22T03:45:29Z)
• Spherical Channels for Modeling Atomic Interactions [11.44635961607696]
  We propose the Spherical Channel Network (SCN) to model atomic energies and forces. SCN is a graph neural network where nodes represent atoms and edges their neighboring atoms. We demonstrate results on the large-scale Open Catalyst 2020 dataset in both energy and force prediction for numerous tasks and metrics.
  arXiv  Detail & Related papers  (2022-06-29T00:03:14Z)
• Directional scrambling of quantum information in helical multiferroics [0.0]
  Local excitations as carriers of quantum information spread out in the system in ways governed by the underlying interaction and symmetry. Character and direction dependence of quantum scrambling can be inferred from the out-of-time-ordered commutators. We study and quantify the directionality of quantum information propagation in oxide-based helical spin systems.
  arXiv  Detail & Related papers  (2021-12-20T17:58:19Z)
• Fast Gravitational Approach for Rigid Point Set Registration with Ordinary Differential Equations [79.71184760864507]
  This article introduces a new physics-based method for rigid point set alignment called Fast Gravitational Approach (FGA) In FGA, the source and target point sets are interpreted as rigid particle swarms with masses interacting in a globally multiply-linked manner while moving in a simulated gravitational force field. We show that the new method class has characteristics not found in previous alignment methods.
  arXiv  Detail & Related papers  (2020-09-28T15:05:39Z)
• A multiconfigurational study of the negatively charged nitrogen-vacancy center in diamond [55.58269472099399]
  Deep defects in wide band gap semiconductors have emerged as leading qubit candidates for realizing quantum sensing and information applications. Here we show that unlike single-particle treatments, the multiconfigurational quantum chemistry methods, traditionally reserved for atoms/molecules, accurately describe the many-body characteristics of the electronic states of these defect centers.
  arXiv  Detail & Related papers  (2020-08-24T01:49:54Z)
• Lorentz Group Equivariant Neural Network for Particle Physics [58.56031187968692]
  We present a neural network architecture that is fully equivariant with respect to transformations under the Lorentz group. For classification tasks in particle physics, we demonstrate that such an equivariant architecture leads to drastically simpler models that have relatively few learnable parameters.
  arXiv  Detail & Related papers  (2020-06-08T17:54:43Z)

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.