Self-Supervised Learning on Molecular Graphs: A Systematic Investigation of Masking Design

TL;DR

This paper systematically investigates masking strategies in self-supervised learning for molecular graphs, revealing that target choice and encoder design are more impactful than complex masking distributions, guiding future SSL development.

Contribution

It introduces a probabilistic framework for evaluating masking strategies and provides empirical insights into their effectiveness in molecular graph SSL.

Findings

Uniform masking performs comparably to complex distributions.

Semantic prediction targets significantly improve downstream tasks.

Graph Transformer encoders benefit more from richer prediction targets.

Abstract

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.