cs.LGNov 17, 2026

Spatially local mini-batching for implicit neural representations

Authors: Minseok Choi, Swarup Dhar, Heejae Kwon, Kookjin Lee, Donsub Rim

Abstract

Neural surrogates based on implicit neural representations (INRs) are typically trained on mini-batches spread over the whole domain, since a generic INR loses accuracy when every point of a batch lies in one small region. We show that low-rank neural representations (LRNRs), INRs whose weight matrices are products of shared bases with coefficients emitted by a small hypernetwork, can be trained with spatially local mini-batches. Each step uses the points of one time snapshot of an adaptive-mesh numerical solution inside one small cell, with snapshots and cells drawn according to the refinement of the mesh. We posit that such training is enabled by the structural property of LRNRs, as a local step reaches the shared bases only through the rr coefficients of each layer, and keeps one cell from rewriting the network. By a Kolmogorov-width argument, we show that when restricted to a small cell, the hidden states of LRNRs lie close to a low-dimensional subspace, which suggests the local steps only affect low-rank subspaces of the shared bases. On six hyperbolic-conservation-law problems with solutions computed using adaptive mesh refinement, one local training recipe reaches 1.1-2.3% relative L2L^2 error with a tenth of the parameters of a dense network of the same width. Finally, we provide a comparison by training a hypernetwork-SIREN, a DeepONet, and a Fourier neural operator (FNO) of the same size with the same local batches and training budget. We observe that these baselines reach 16-78% where the LRNR reaches 4.9%. Interestingly, we also observe that local training has a significant effect on the per-cell low-dimensional structure even in a wide generic hypernetwork-SIREN.

Figures & tables

Appendix figures & tables24 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Refined Grid Sampling for Efficient Implicit Neural Representations of Large-Scale Scientific Data

    Oct 9, 2026Guang Zhao, Xihaier Luo, Sean McCue +5Implicit Neural RepresentationsAdaptive Sampling

  2. Refine Now, Query Fast: A Decoupled Refinement Paradigm for Implicit Neural Fields

    Feb 16, 2026Tianyu Xiong, Skylar Wurster, Han-Wei ShenImplicit Neural RepresentationsNeural Surrogate Modeling