Weisfeiler and Leman Go Measurement Modeling: Probing the Validity of the WL Test
This work critiques a foundational tool in graph machine learning, highlighting its limitations for practitioners in developing and communicating expressive power.
The paper uncovers misalignments between practitioners' conceptualizations of expressive power and the k-WL test, revealing that k-WL does not guarantee isometry, can be irrelevant to real-world tasks, and may not promote generalization or trustworthiness.
The expressive power of graph neural networks is usually measured by comparing how many pairs of graphs or nodes an architecture can possibly distinguish as non-isomorphic to those distinguishable by the $k$-dimensional Weisfeiler-Leman ($k$-WL) test. In this paper, we uncover misalignments between graph machine learning practitioners' conceptualizations of expressive power and $k$-WL through a systematic analysis of the reliability and validity of $k$-WL. We conduct a survey ($n = 18$) of practitioners to surface their conceptualizations of expressive power and their assumptions about $k$-WL. In contrast to practitioners' beliefs, our analysis (which draws from graph theory and benchmark auditing) reveals that $k$-WL does not guarantee isometry, can be irrelevant to real-world graph tasks, and may not promote generalization or trustworthiness. We argue for extensional definitions and measurement of expressive power based on benchmarks. We further contribute guiding questions for constructing such benchmarks, which is critical for graph machine learning practitioners to develop and transparently communicate our understandings of expressive power.