Low Dimensional Sampling under Reconstructed Constraints
Abstract
We study sampling from a distribution supported on an unknown compact d-dimensional C2 manifold M⊂ℝD, observed only through i.i.d. uniform points from M. We reconstruct the constraint using an adaptive local-convex-hull estimator and target an ambient distribution penalized by squared distance to the reconstruction. Although the reconstructed set may be nonsmooth or fail to be a manifold, its Hausdorff accuracy alone suffices to control the Wasserstein error. We quantify the tradeoff between reconstruction accuracy and penalty strength and show that optimal tuning gives an error proportional to the square root of the Hausdorff error. A planar example proves that this dependence is sharp for the proposed scheme. For a d dimensional constraint reconstructed from N observations, the error becomes 𝒪( (N/N)1/d). Finally, we prove uniform geometric ergodicity of a Gaussian random-walk Metropolis--Hastings sampler and combine reconstruction, approximation, and mixing into an explicit finite-time guarantee.
Related
- Inverse domain uncertainty quantification in hemodynamics: sampling virtual patient cohorts under constraints
- Sampling with Autoregressive Clock Jitter: Infinite-Dimensional Stability Failure and Finite-Dimensional Conditioning
- Sharp dimensional analysis of midpoint methods for Langevin sampling
- Dynamic Control of a Make-to-Order Manufacturing System under Throughput Time Constraints: An Effective Computational Method in the High-Dimensional Case
- Any-Dimensional Learning by Sampling
BibTeX
@misc{banerjee2026dimensional,
title = {{Low Dimensional Sampling under Reconstructed Constraints}},
author = {Imon Banerjee and Riddhiman Bhattacharyya},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.14837},
eprint = {2609.14837},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.14837},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Low Dimensional Sampling under Reconstructed Constraints
AU - Imon Banerjee
AU - Riddhiman Bhattacharyya
PY - 2026
DA - 2026/09/13
PB - arXiv
DO - 10.48550/arxiv.2609.14837
UR - https://arxiv.org/abs/2609.14837
AB - We study sampling from a distribution supported on an unknown compact d-dimensional C^2 manifold M⊂ℝ^D, observed only through i.i.d. uniform points from M. We reconstruct the constraint using an adaptive local-convex-hull estimator and target an ambient distribution penalized by squared distance to the reconstruction. Although the reconstructed set may be nonsmooth or fail to be a manifold, its Hausdorff accuracy alone suffices to control the Wasserstein error. We quantify the tradeoff between reconstruction accuracy and penalty strength and show that optimal tuning gives an error proportional to the square root of the Hausdorff error. A planar example proves that this dependence is sharp for the proposed scheme. For a d dimensional constraint reconstructed from N observations, the error becomes 𝒪( (N/N)^1/d). Finally, we prove uniform geometric ergodicity of a Gaussian random-walk Metropolis--Hastings sampler and combine reconstruction, approximation, and mixing into an explicit finite-time guarantee.
ER -