Model-Based Learning of Latent Euclidean Embedding from Relational Data
Abstract
Visualizing and interpreting the dependence structure in relational data often benefits from geometric representations via low-dimensional Euclidean embeddings. However, existing approaches are typically restricted to specific response types, rely on burdensome high-dimensional nonconvex optimization, or lack uniform theoretical guarantees for simultaneous coordinate recovery. To address these limitations, we introduce a unified, distance-based latent space framework for both unipartite and bipartite relational data---accommodating continuous, binary, ordinal, and count-valued observations---via a modular two-stage procedure that decouples signal estimation from geometric recovery. First, we estimate the low-rank signal matrix using spectral denoising. For models with nonlinear link functions, we further apply one-step row- and column-wise refinement to achieve uniform entrywise accuracy. Second, we recover the latent configurations through direct spectral analysis for unipartite data and a low-dimensional alignment optimization for bipartite data. We establish uniform recovery guarantees that characterize the propagation of error from signal estimation to the embedding space. To illustrate the scalability and interpretability of our approach, we apply the proposed method to two real-world datasets: a statistician citation network and European Parliament voting records.
Related
- Personalised federated learning for Riemannian and Euclidean EEG decoding
- Bounded Adjustment with Reliability-Guided Embedding for Imbalanced Learning with Noisy Labels
- Provable Guarantees and Efficient Learning of Structural Equation Models with Latent Confounders
- Different representation learning objectives recover distinct latent structures from the same psychometric data
- Learning Stabilizing Latent Feedback Controllers via Multi-Step Backward Losses
BibTeX
@misc{chen2026modelbased,
title = {{Model-Based Learning of Latent Euclidean Embedding from Relational Data}},
author = {Yinghang Chen and Yunxiao Chen and Haoran Zhang},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.00765},
eprint = {2611.00765},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.00765},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Model-Based Learning of Latent Euclidean Embedding from Relational Data
AU - Yinghang Chen
AU - Yunxiao Chen
AU - Haoran Zhang
PY - 2026
DA - 2026/11/01
PB - arXiv
DO - 10.48550/arxiv.2611.00765
UR - https://arxiv.org/abs/2611.00765
AB - Visualizing and interpreting the dependence structure in relational data often benefits from geometric representations via low-dimensional Euclidean embeddings. However, existing approaches are typically restricted to specific response types, rely on burdensome high-dimensional nonconvex optimization, or lack uniform theoretical guarantees for simultaneous coordinate recovery. To address these limitations, we introduce a unified, distance-based latent space framework for both unipartite and bipartite relational data---accommodating continuous, binary, ordinal, and count-valued observations---via a modular two-stage procedure that decouples signal estimation from geometric recovery. First, we estimate the low-rank signal matrix using spectral denoising. For models with nonlinear link functions, we further apply one-step row- and column-wise refinement to achieve uniform entrywise accuracy. Second, we recover the latent configurations through direct spectral analysis for unipartite data and a low-dimensional alignment optimization for bipartite data. We establish uniform recovery guarantees that characterize the propagation of error from signal estimation to the embedding space. To illustrate the scalability and interpretability of our approach, we apply the proposed method to two real-world datasets: a statistician citation network and European Parliament voting records.
ER -