Consistent Spatial Clustering of Complex Objects via Bregman Geometry
Abstract
Spatial clustering increasingly involves complex objects, including probability distributions, networks, and structured matrices, for which conventional parametric likelihoods may be difficult to specify. We construct a generalized likelihood by exponentiating a cluster-specific Bregman loss on fixed representations of the responses, avoiding the need to specify a parametric sampling model for each response class. Each cluster is associated with an unknown representative, and the generalized likelihood measures within-cluster homogeneity through the Bregman divergence between the represented observations and that representative. A prior family matched to the Bregman geometry yields conjugate generalized posterior updates and permits exact marginalization of the cluster-specific representatives. The resulting collapsed scores combine within-cluster Bregman dispersion with uncertainty in the representatives, while a spanning-tree partition prior restricts posterior partition support to spatially contiguous clusters. This collapsed representation also leads to a tractable MCMC algorithm for posterior computation. The framework accommodates multiple complex-object response classes within a common inferential construction. We establish posterior consistency for the spatial partition and cluster-specific representatives under infill-domain asymptotics. Simulation studies with different complex-object responses demonstrate accurate recovery of spatial partitions and cluster-specific representatives. We further analyze racial-composition distributions across spatial units in Houston, illustrating geographically coherent clusters with distinct demographic compositions.
Related
- DISCCO: Distance-Based Bayesian Spatial Clustering of Complex Objects with Node-Frailty Centrality
- A spectral clustering-type algorithm for the consistent estimation of the Hurst distribution in moderately high dimensions
- Additive Nonparametric Regression with Spatial and Network Objects
- ON HOPFIAN AND CO-HOPFIAN OBJECTS IN AN ABELIAN CATEGORY AND IN THE CATEGORY OF ITS COMPLEX SEQUENCES OF MORPHISMS
- A method for detecting small objects in remote sensing images based on feature enhancement and scale-adaptive optimization in complex scenarios
BibTeX
@misc{bhattacharyya2026consistent,
title = {{Consistent Spatial Clustering of Complex Objects via Bregman Geometry}},
author = {Srijato Bhattacharyya and Huiyan Sang},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.04166},
eprint = {2610.04166},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.04166},
}
RIS
TY - UNPB
TI - Consistent Spatial Clustering of Complex Objects via Bregman Geometry
AU - Srijato Bhattacharyya
AU - Huiyan Sang
PY - 2026
DA - 2026/10/03
PB - arXiv
DO - 10.48550/arxiv.2610.04166
UR - https://arxiv.org/abs/2610.04166
AB - Spatial clustering increasingly involves complex objects, including probability distributions, networks, and structured matrices, for which conventional parametric likelihoods may be difficult to specify. We construct a generalized likelihood by exponentiating a cluster-specific Bregman loss on fixed representations of the responses, avoiding the need to specify a parametric sampling model for each response class. Each cluster is associated with an unknown representative, and the generalized likelihood measures within-cluster homogeneity through the Bregman divergence between the represented observations and that representative. A prior family matched to the Bregman geometry yields conjugate generalized posterior updates and permits exact marginalization of the cluster-specific representatives. The resulting collapsed scores combine within-cluster Bregman dispersion with uncertainty in the representatives, while a spanning-tree partition prior restricts posterior partition support to spatially contiguous clusters. This collapsed representation also leads to a tractable MCMC algorithm for posterior computation. The framework accommodates multiple complex-object response classes within a common inferential construction. We establish posterior consistency for the spatial partition and cluster-specific representatives under infill-domain asymptotics. Simulation studies with different complex-object responses demonstrate accurate recovery of spatial partitions and cluster-specific representatives. We further analyze racial-composition distributions across spatial units in Houston, illustrating geographically coherent clusters with distinct demographic compositions.
ER -