Saved in:
Bibliographic Details
Main Authors: Carlon, May Kristine Jonson, Noe, Su Myat, Wang, Haojiong, Kuniyoshi, Yasuo
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.03217
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912871815839744
author Carlon, May Kristine Jonson
Noe, Su Myat
Wang, Haojiong
Kuniyoshi, Yasuo
author_facet Carlon, May Kristine Jonson
Noe, Su Myat
Wang, Haojiong
Kuniyoshi, Yasuo
contents Understanding how local interactions give rise to global brain organization requires models that can represent information across multiple scales. We introduce a hierarchical self-supervised learning (SSL) framework that jointly learns node-, edge-, and graph-level embeddings, inspired by multimodal neuroimaging. We construct a controllable synthetic benchmark mimicking the topological properties of connectomes. Our four-stage evaluation protocol reveals a critical failure: the invariance-based SSL model is fundamentally misaligned with the benchmark's topological properties and is catastrophically outperformed by classical, topology-aware heuristics. Ablations confirm an objective mismatch: SSL objectives designed to be invariant to topological perturbations learn to ignore the very community structure that classical methods exploit. Our results expose a fundamental pitfall in applying generic graph SSL to connectome-like data. We present this framework as a cautionary case study, highlighting the need for new, topology-aware SSL objectives for neuro-AI research that explicitly reward the preservation of structure (e.g., modularity or motifs).
format Preprint
id arxiv_https___arxiv_org_abs_2602_03217
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Topology Matters: A Cautionary Case Study of Graph SSL on Neuro-Inspired Benchmarks
Carlon, May Kristine Jonson
Noe, Su Myat
Wang, Haojiong
Kuniyoshi, Yasuo
Machine Learning
Artificial Intelligence
Understanding how local interactions give rise to global brain organization requires models that can represent information across multiple scales. We introduce a hierarchical self-supervised learning (SSL) framework that jointly learns node-, edge-, and graph-level embeddings, inspired by multimodal neuroimaging. We construct a controllable synthetic benchmark mimicking the topological properties of connectomes. Our four-stage evaluation protocol reveals a critical failure: the invariance-based SSL model is fundamentally misaligned with the benchmark's topological properties and is catastrophically outperformed by classical, topology-aware heuristics. Ablations confirm an objective mismatch: SSL objectives designed to be invariant to topological perturbations learn to ignore the very community structure that classical methods exploit. Our results expose a fundamental pitfall in applying generic graph SSL to connectome-like data. We present this framework as a cautionary case study, highlighting the need for new, topology-aware SSL objectives for neuro-AI research that explicitly reward the preservation of structure (e.g., modularity or motifs).
title Topology Matters: A Cautionary Case Study of Graph SSL on Neuro-Inspired Benchmarks
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2602.03217