Learning Approximate Isometric Embeddings for Efficient Template Placement
Abstract
The sphere covering problem has broad applications in science however, exact solutions are computationally infeasible on curved, high-dimensional manifolds because distances must be evaluated using a position-dependent metric. We present isobank, a normalizing flow that learns an approximately isometric coordinate transformation, such that Euclidean distances in the latent space approximate metric-induced proper distances in the physical parameter space. This enables template placement using Euclidean distance checks in the latent space, avoiding repeated metric evaluations during placement. We evaluate isobank in the context of gravitational wave template bank construction, where the sphere covering problem corresponds to the template placement problem. On a five-dimensional eccentric binary parameter space, isobank reaches 97\% coverage at a maximum mismatch of μ=0.03 with 23\% fewer templates than the state-of-the-art stochastic placement code mbank, which reaches 94\% coverage. Coverage is measured on independent injections drawn uniformly over the masses, spins and eccentricity, using the exact waveform match rather than the metric approximation. Training the flow takes about 2.5h and placing the bank under 1h on 32 CPU cores. isobank is publicly available on Github https://github.com/Alexandra-Wernersson/isobank
Related
- An Approximate Bayesian Deep Learning Approach for Uncertainty-aware Differential Emission Measure Estimates in the Solar Corona from the SDO
- Spin Spherical Fourier Neural Operators: Efficient Equivariant Learning for Vector and Spin Fields on the Sphere
- CausticFlow: An Efficient Machine Learning Framework Combining Neural Differential Equations and Normalizing Flows for Binary Microlensing Parameter Inference
- CausticFlow: An Efficient Machine Learning Framework Combining Neural Differential Equations and Normalizing Flows for Binary Microlensing Parameter Inference
- Morphology classification for galaxies in the Kilo Degree Survey using a label-efficient self-supervised learning framework
BibTeX
@misc{wernersson2026learning,
title = {{Learning Approximate Isometric Embeddings for Efficient Template Placement}},
author = {Alexandra Wernersson and Jessica Irwin and Melissa Lopez and Sarah Caudill},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.02448},
eprint = {2610.02448},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.02448},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Learning Approximate Isometric Embeddings for Efficient Template Placement
AU - Alexandra Wernersson
AU - Jessica Irwin
AU - Melissa Lopez
AU - Sarah Caudill
PY - 2026
DA - 2026/10/01
PB - arXiv
DO - 10.48550/arxiv.2610.02448
UR - https://arxiv.org/abs/2610.02448
AB - The sphere covering problem has broad applications in science however, exact solutions are computationally infeasible on curved, high-dimensional manifolds because distances must be evaluated using a position-dependent metric. We present isobank, a normalizing flow that learns an approximately isometric coordinate transformation, such that Euclidean distances in the latent space approximate metric-induced proper distances in the physical parameter space. This enables template placement using Euclidean distance checks in the latent space, avoiding repeated metric evaluations during placement. We evaluate isobank in the context of gravitational wave template bank construction, where the sphere covering problem corresponds to the template placement problem. On a five-dimensional eccentric binary parameter space, isobank reaches 97\% coverage at a maximum mismatch of μ=0.03 with 23\% fewer templates than the state-of-the-art stochastic placement code mbank, which reaches 94\% coverage. Coverage is measured on independent injections drawn uniformly over the masses, spins and eccentricity, using the exact waveform match rather than the metric approximation. Training the flow takes about 2.5h and placing the bank under 1h on 32 CPU cores. isobank is publicly available on Github https://github.com/Alexandra-Wernersson/isobank
ER -