SH-WRNN: Implicit Spherical Harmonics Weight Field Routing Neural Networks for Asymmetric Edge Intelligence

• URL: http://arxiv.org/abs/2609.14614v1
• Date: Sun, 13 Sep 2026 15:44:47 GMT
• Title: SH-WRNN: Implicit Spherical Harmonics Weight Field Routing Neural Networks for Asymmetric Edge Intelligence
• Abstract summary: We introduce the Implicit Spherical Harmonics Weight Field Routing Neural Network (SH-WRNN)<n>When retrieving the weight matrix of the current layer, connection parameters are localized using latitude and longitude.<n>SH-WRNN achieves robust accuracies of 91.05% and 81.54% within a single training epoch.
• Score: 0.5500249707065664
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Deep learning architectures remain rigidly built upon traditional fully connected layers. While networks scale up, few challenge this foundational root. In this work, we reshape this paradigm by transforming the core synapse weight matrix from static, discrete parameters into a differentiable, continuous field governed by spherical harmonics functions. We introduce the Implicit Spherical Harmonics Weight Field Routing Neural Network (SH-WRNN), which constrains weight matrices within a continuous parametric field instead of optimizing millions of localized discrete weights. When retrieving the weight matrix of the current layer, connection parameters are localized using latitude and longitude on a rectangular plane mapped from the continuous field. The latitudinal coordinate is specified by activated neurons from the previous layer, while the longitudinal coordinate is determined by keys generated from previous layer activations via matrix multiplication. By evaluating intersections on this map, the network dynamically extracts its connection weights on-the-fly. Empirical validation on MNIST demonstrates that under compact configurations of (32, 10, 10) and (32, 3, 10), SH-WRNN achieves robust accuracies of 91.05% and 81.54% within a single training epoch. Furthermore, we propose an asymmetric Surface Baking scheme. Upon convergence, the continuous weight field is baked once into a static parametric surface. By eliminating analytical spherical harmonics calculations during inference and reducing dynamic matrix extraction to high-speed localized memory slicing, this scheme achieves asymmetric algorithmic acceleration with negligible accuracy degradation. This paradigm shift bypasses GPU memory-bandwidth monopolies, opening a novel path to reshape the advantages of CPU computing. Code is available at https://github.com/jzb1111/SphericalHarmonyRoutedNeuralNetWork.

Related papers

• Emulating Cosmic Structure Formation with a Lagrangian Neural Cellular Automaton [0.0]
  We introduce a hybrid deep learning framework that can be applied to structure formation as a local, iterative dynamical process on a comoving lattice.<n>By training the network to learn only the textitresidual displacement corrections to the Zeldovich approximation, we achieve high-fidelity emulation of the non-linear physics.<n>Our trained model supports percent-level precision in the power and cross spectra well into the non-linear regime.
  arXiv  Detail & Related papers  (2026-07-29T18:00:01Z)
• Equivariant Continuous Normalizing Flows with Offline Sampling for Fermionic Ground State Estimation [0.0]
  Flow is implemented as a permutation-equivariant neural ODE.<n>Training runs on systems of harmonically trapped spinless electrons demonstrate ground-state energies below CISD reference values.<n>Scaling experiments demonstrate near-ideal strong scaling from 1 to 128 NVIDIA A100s using 32 GPU nodes of NERSC's Perlmutter supercomputer.
  arXiv  Detail & Related papers  (2026-07-20T20:17:13Z)
• Machine Learning-Based Reconstruction for Resistive Silicon Sensors [53.365453214937496]
  We study machine-learning-based reconstruction and compression for resistive silicon sensors.<n>We use full-waveform information from correlated pads to regularise the reconstruction and extract spatial information.<n>We first introduce recurrent neural network models based on LSTM layers, which provide a proof-of-concept implementation for full-waveform reconstruction.
  arXiv  Detail & Related papers  (2026-07-13T14:03:26Z)
• Non-linear mechanical field reconstruction coupling recurrent neural networks with physics-informed graph neural networks [0.6978367196609415]
  We propose a coupled LSTMGNN framework that links the temporal and spatial aspects of local stress field reconstruction.<n>We introduce a relative weighting strategy with linear warm-up to balance the data-driven reconstruction loss and a discrete divergence-based equilibrium penalty.<n>The model achieves three orders of magnitude speedup over finite element simulations and generalizes to loading sequences twice the training length, with 1.9% cumulative error.
  arXiv  Detail & Related papers  (2026-06-09T14:20:39Z)
• NNiT: Width-Agnostic Neural Network Generation with Structurally Aligned Weight Spaces [15.631276865948097]
  We introduce Neural Network Diffusion Transformers (NNiTs), which generate weights in a width-agnostic manner.<n>On ManiSkill3 robotics tasks, NNiT achieves >85% success on architecture topologies unseen during training, while baseline approaches fail to generalize.
  arXiv  Detail & Related papers  (2026-02-26T20:47:30Z)
• Point Cloud Denoising With Fine-Granularity Dynamic Graph Convolutional Networks [58.050130177241186]
  Noise perturbations often corrupt 3-D point clouds, hindering downstream tasks such as surface reconstruction, rendering, and further processing. This paper introduces finegranularity dynamic graph convolutional networks called GDGCN, a novel approach to denoising in 3-D point clouds.
  arXiv  Detail & Related papers  (2024-11-21T14:19:32Z)
• Enhancing lattice kinetic schemes for fluid dynamics with Lattice-Equivariant Neural Networks [79.16635054977068]
  We present a new class of equivariant neural networks, dubbed Lattice-Equivariant Neural Networks (LENNs) Our approach develops within a recently introduced framework aimed at learning neural network-based surrogate models Lattice Boltzmann collision operators. Our work opens towards practical utilization of machine learning-augmented Lattice Boltzmann CFD in real-world simulations.
  arXiv  Detail & Related papers  (2024-05-22T17:23:15Z)
• Dynamic Graph: Learning Instance-aware Connectivity for Neural Networks [78.65792427542672]
  Dynamic Graph Network (DG-Net) is a complete directed acyclic graph, where the nodes represent convolutional blocks and the edges represent connection paths. Instead of using the same path of the network, DG-Net aggregates features dynamically in each node, which allows the network to have more representation ability.
  arXiv  Detail & Related papers  (2020-10-02T16:50:26Z)
• Accurate and efficient Simulation of very high-dimensional Neural Mass Models with distributed-delay Connectome Tensors [0.23453441553817037]
  This paper introduces methods that efficiently integrates any high-dimensional Neural Mass Models (NMMs) specified by two essential components. The first is the set of nonlinear Random Differential Equations of the dynamics of each neural mass. The second is the highly sparse three-dimensional Connectome (CT) that encodes the strength of the connections and the delays of information transfer along the axons of each connection.
  arXiv  Detail & Related papers  (2020-09-16T05:55:17Z)
• From deep to Shallow: Equivalent Forms of Deep Networks in Reproducing Kernel Krein Space and Indefinite Support Vector Machines [63.011641517977644]
  We take a deep network and convert it to an equivalent (indefinite) kernel machine. We then investigate the implications of this transformation for capacity control and uniform convergence. Finally, we analyse the sparsity properties of the flat representation, showing that the flat weights are (effectively) Lp-"norm" regularised with 0p1.
  arXiv  Detail & Related papers  (2020-07-15T03:21:35Z)
• Neural Subdivision [58.97214948753937]
  This paper introduces Neural Subdivision, a novel framework for data-driven coarseto-fine geometry modeling. We optimize for the same set of network weights across all local mesh patches, thus providing an architecture that is not constrained to a specific input mesh, fixed genus, or category. We demonstrate that even when trained on a single high-resolution mesh our method generates reasonable subdivisions for novel shapes.
  arXiv  Detail & Related papers  (2020-05-04T20:03:21Z)
• Nonlinear system identification with regularized Tensor Network B-splines [2.817412580574242]
  The TNBS-NARX model is validated through the identification of the cascaded watertank benchmark nonlinear system. It achieves state-of-the-art performance while identifying a 16-dimensional B-spline surface in 4 seconds on a standard desktop computer.
  arXiv  Detail & Related papers  (2020-03-17T09:22:20Z)

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.