Aggregate to Adapt: Node-Centric Aggregation for Multi-Source-Free Graph Domain Adaptation

TL;DR

This paper introduces GraphATA, a novel node-centric method for multi-source-free unsupervised graph domain adaptation that dynamically aggregates source models at the node level, improving transfer performance without source data.

Contribution

Proposes GraphATA, a new node-level adaptation approach that aggregates source models dynamically, addressing multi-source-free graph domain adaptation challenges.

Findings

GraphATA outperforms recent baselines on multiple datasets.

Node-level aggregation improves adaptation effectiveness.

Method generalizes to various model and layer configurations.

Abstract

Unsupervised graph domain adaptation (UGDA) focuses on transferring knowledge from labeled source graph to unlabeled target graph under domain discrepancies. Most existing UGDA methods are designed to adapt information from a single source domain, which cannot effectively exploit the complementary knowledge from multiple source domains. Furthermore, their assumptions that the labeled source graphs are accessible throughout the training procedure might not be practical due to privacy, regulation, and storage concerns. In this paper, we investigate multi-source-free unsupervised graph domain adaptation, i.e., adapting knowledge from multiple source domains to an unlabeled target domain without utilizing labeled source graphs but relying solely on source pre-trained models. Unlike previous multi-source domain adaptation approaches that aggregate predictions at model level, we introduce a novel model named GraphATA which conducts adaptation at node granularity. Specifically, we parameterize each node with its own graph convolutional matrix by automatically aggregating weight matrices from multiple source models according to its local context, thus realizing dynamic adaptation over graph structured data. We also demonstrate the capability of GraphATA to generalize to both model-centric and layer-centric methods. Comprehensive experiments on various public datasets show that our GraphATA can consistently surpass recent state-of-the-art baselines with different gains.