GALA: Graph Diffusion-based Alignment with Jigsaw for Source-free Domain Adaptation

Abstract

Source-free domain adaptation is a crucial machine learning topic, as it contains numerous applications in the real world, particularly with respect to data privacy. Existing approaches predominantly focus on Euclidean data, such as images and videos, while the exploration of non-Euclidean graph data remains scarce. Recent graph neural network (GNN) approaches can suffer from serious performance decline due to domain shift and label scarcity in source-free adaptation scenarios. In this study, we propose a novel method named Graph Diffusion-based Alignment with Jigsaw (GALA), tailored for source-free graph domain adaptation. To achieve domain alignment, GALA employs a graph diffusion model to reconstruct source-style graphs from target data. Specifically, a score-based graph diffusion model is trained using source graphs to learn the generative source styles. Then, we introduce perturbations to target graphs via a stochastic differential equation instead of sampling from a prior, followed by the reverse process to reconstruct source-style graphs. We feed the source-style graphs into an off-the-shelf GNN and introduce class-specific thresholds with curriculum learning, which can generate accurate and unbiased pseudo-labels for target graphs. Moreover, we develop a simple yet effective graph-mixing strategy named graph jigsaw to combine confident graphs and unconfident graphs, which can enhance generalization capabilities and robustness via consistency learning. Extensive experiments on benchmark datasets validate the effectiveness of GALA.

IEEE TPAMI

References in corpus (23)

• Fast unfolding of communities in large networks
• A Comprehensive Survey on Graph Neural Networks
• Semi-Supervised Classification with Graph Convolutional Networks
• Learning Transferable Features with Deep Adaptation Networks
• FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence
• Score-Based Generative Modeling through Stochastic Differential Equations
• Hierarchical Graph Representation Learning with Differentiable Pooling
• FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labeling
• Asymmetric Tri-training for Unsupervised Domain Adaptation
• Self-Attention Graph Pooling
• A Comprehensive Survey on Deep Graph Representation Learning
• SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations
• Improving robustness against common corruptions by covariate shift adaptation
• Evaluating Prediction-Time Batch Normalization for Robustness under Covariate Shift
• Discovering Invariant Rationales for Graph Neural Networks
• G-Mixup: Graph Data Augmentation for Graph Classification
• Semi-Supervised Hierarchical Graph Classification
• KGNN: Harnessing Kernel-based Networks for Semi-supervised Graph Classification
• Handling Distribution Shifts on Graphs: An Invariance Perspective
• SizeShiftReg: a Regularization Method for Improving Size-Generalization in Graph Neural Networks
• CoCo: A Coupled Contrastive Framework for Unsupervised Domain Adaptive Graph Classification
• Structural Re-weighting Improves Graph Domain Adaptation
• GraphGDP: Generative Diffusion Processes for Permutation Invariant Graph Generation

Cited by in corpus (2)

• Graph Domain Adaptation: Challenges, Progress and Prospects
• Cross-Domain Diffusion with Progressive Alignment for Efficient Adaptive Retrieval