Topology Matters: A Cautionary Case Study of Graph SSL on Neuro-Inspired Benchmarks

TL;DR

This study critically evaluates a hierarchical self-supervised learning framework on neuro-inspired benchmarks, revealing that invariance-based SSL models fail to capture essential topological features, unlike classical topology-aware methods.

Contribution

The paper introduces a synthetic connectome benchmark and demonstrates that standard SSL objectives are misaligned with topological properties, emphasizing the need for topology-aware SSL approaches.

Findings

SSL models ignore community structure in connectome-like data

Classical topology-aware heuristics outperform SSL in the benchmark

Invariance-based SSL objectives lead to objective mismatch and failure

Abstract

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).