Synapse
⌘+K
Synapse
PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
October 20, 2025Open Access

Graph Distribution-valued Signals: A Wasserstein Space Perspective

View Full Paper
Ask AI
Bookmark
Share

Authors

YZYanan ZhaoFJFeng JiXJXingchao Jian

Discussion

Loading...

Member takes

Overview

Novel framework models signals as graph distribution-valued signals, suggesting improved handling of uncertainty in predictions.

Key Points

  • This framework introduces graph distribution-valued signals, enhancing traditional graph signal processing approaches.
  • By representing signals as distributions, GDSs allow for better handling of uncertainty and stochasticity in graph filtering.
  • A systematic mapping connects traditional graph signal processing concepts to the new distribution-based framework.
  • Experimental validation shows the framework effectively supports graph filter learning for various prediction tasks.

Cite This Study

Zhao et al. (2025) studied this question.

synapsesocial.com/papers/68f5fcd68d54a28a75cf1ed6https://doi.org/10.48550/arxiv.2509.25802
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1On Generalized Signature Graphs2024 · 1 citations
  2. 2Graphon Signal Processing for Spiking and Biological Neural Networks.2026
  3. 3Unified Fourier bases for GSP on stochastic block model graphs2024
  4. 4Fast High-Quality Directed Graph Learning2024
  5. 5Reconstruction of Graph Signals on Complex Manifolds with Kernel Methods2025