Guardado en:
Detalles Bibliográficos
Autores principales: Chuang, Gabriel, Chaintreau, Augustin
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2509.12396
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866918313960931328
author Chuang, Gabriel
Chaintreau, Augustin
author_facet Chuang, Gabriel
Chaintreau, Augustin
contents An extensive line of work studies fairness interventions for network embeddings, but less is known about their baseline behavior. In this work, we ask: how do baseline embeddings (without fairness interventions) produce disparate effects at the representation level? We analyze the asymptotic behavior of low-dimensional embeddings on stochastic block model (SBM) graphs, which encode both homophily and group structure. We characterize exact conditions under which embeddings cause information loss, showing that the amount of information loss depends directly on the graph's density and assortativity. Notably, very different graphs can produce identical embeddings in the limit, and this non-invertibility disproportionately affects smaller and sparser communities. As a result, simple downstream tasks, such as link prediction, introduce higher error rates for these communities, helping explain disparities widely observed in practice.
format Preprint
id arxiv_https___arxiv_org_abs_2509_12396
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Information Loss and Disparate Effects in Network Embeddings
Chuang, Gabriel
Chaintreau, Augustin
Social and Information Networks
Machine Learning
An extensive line of work studies fairness interventions for network embeddings, but less is known about their baseline behavior. In this work, we ask: how do baseline embeddings (without fairness interventions) produce disparate effects at the representation level? We analyze the asymptotic behavior of low-dimensional embeddings on stochastic block model (SBM) graphs, which encode both homophily and group structure. We characterize exact conditions under which embeddings cause information loss, showing that the amount of information loss depends directly on the graph's density and assortativity. Notably, very different graphs can produce identical embeddings in the limit, and this non-invertibility disproportionately affects smaller and sparser communities. As a result, simple downstream tasks, such as link prediction, introduce higher error rates for these communities, helping explain disparities widely observed in practice.
title Information Loss and Disparate Effects in Network Embeddings
topic Social and Information Networks
Machine Learning
url https://arxiv.org/abs/2509.12396