Distributional Stability of Tangent-Linearized Gaussian Inference on Smooth Manifolds

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Seo, Junghoon, Lee, Hakjin, Sim, Jaehoon
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915965181100032
author Seo, Junghoon
Lee, Hakjin
Sim, Jaehoon
author_facet Seo, Junghoon
Lee, Hakjin
Sim, Jaehoon
contents Gaussian inference on smooth manifolds is central to robotics, but exact marginalization and conditioning are generally non-Gaussian and geometry-dependent. We study tangent-linearized Gaussian inference and derive explicit non-asymptotic $W_2$ stability bounds for projection marginalization and surface-measure conditioning. The bounds separate local second-order geometric distortion from nonlocal tail leakage and, for Gaussian inputs, yield closed-form diagnostics from $(μ,Σ)$ and curvature/reach surrogates. Circle and planar-pushing experiments validate the predicted calibration transition near $\sqrt{\|Σ\|_{\mathrm{op}}}/R\approx 1/6$ and indicate that normal-direction uncertainty is the dominant failure mode when locality breaks. These diagnostics provide practical triggers for switching from single-chart linearization to multi-chart or sample-based manifold inference. Code and Jupyter notebooks are available at https://github.com/mikigom/StabilityTLGaussian.
format Preprint
id arxiv_https___arxiv_org_abs_2602_19179
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Distributional Stability of Tangent-Linearized Gaussian Inference on Smooth Manifolds
Seo, Junghoon
Lee, Hakjin
Sim, Jaehoon
Robotics
Systems and Control
Gaussian inference on smooth manifolds is central to robotics, but exact marginalization and conditioning are generally non-Gaussian and geometry-dependent. We study tangent-linearized Gaussian inference and derive explicit non-asymptotic $W_2$ stability bounds for projection marginalization and surface-measure conditioning. The bounds separate local second-order geometric distortion from nonlocal tail leakage and, for Gaussian inputs, yield closed-form diagnostics from $(μ,Σ)$ and curvature/reach surrogates. Circle and planar-pushing experiments validate the predicted calibration transition near $\sqrt{\|Σ\|_{\mathrm{op}}}/R\approx 1/6$ and indicate that normal-direction uncertainty is the dominant failure mode when locality breaks. These diagnostics provide practical triggers for switching from single-chart linearization to multi-chart or sample-based manifold inference. Code and Jupyter notebooks are available at https://github.com/mikigom/StabilityTLGaussian.
title Distributional Stability of Tangent-Linearized Gaussian Inference on Smooth Manifolds
topic Robotics
Systems and Control
url https://arxiv.org/abs/2602.19179