Self-Supervised Learning on Molecular Graphs: A Systematic Investigation of Masking Design
This work provides practical guidance for researchers developing self-supervised learning methods in molecular graph representation learning, though it is incremental as it refines existing approaches rather than introducing a new paradigm.
The paper tackled the problem of unclear effectiveness in masking-based self-supervised learning for molecular graphs by systematically evaluating design choices, finding that uniform masking often suffices and that prediction target and encoder synergy are more critical, with specific improvements noted for richer targets and Graph Transformers.
Self-supervised learning (SSL) plays a central role in molecular representation learning. Yet, many recent innovations in masking-based pretraining are introduced as heuristics and lack principled evaluation, obscuring which design choices are genuinely effective. This work cast the entire pretrain-finetune workflow into a unified probabilistic framework, enabling a transparent comparison and deeper understanding of masking strategies. Building on this formalism, we conduct a controlled study of three core design dimensions: masking distribution, prediction target, and encoder architecture, under rigorously controlled settings. We further employ information-theoretic measures to assess the informativeness of pretraining signals and connect them to empirically benchmarked downstream performance. Our findings reveal a surprising insight: sophisticated masking distributions offer no consistent benefit over uniform sampling for common node-level prediction tasks. Instead, the choice of prediction target and its synergy with the encoder architecture are far more critical. Specifically, shifting to semantically richer targets yields substantial downstream improvements, particularly when paired with expressive Graph Transformer encoders. These insights offer practical guidance for developing more effective SSL methods for molecular graphs.