HetGPT: Harnessing the Power of Prompt Tuning in Pre-Trained Heterogeneous Graph Neural Networks

📅 2023-10-23
🏛️ The Web Conference
📈 Citations: 9
Influential: 3
📄 PDF
🤖 AI Summary
This work addresses negative transfer in pre-trained heterogeneous graph neural networks (HGNNs) for semi-supervised node classification, caused by objective mismatch between pre-training and downstream tasks. We propose a novel “pre-training + prompting” paradigm. Our method introduces: (1) the first heterogeneous-graph-aware prompt function, jointly modeling virtual class prompts and heterogeneous feature prompts; and (2) a multi-view neighbor aggregation mechanism, enabling the first general adaptation of prompt tuning to HGNNs. Crucially, our approach requires no fine-tuning of model parameters—only lightweight prompt modules are optimized. Evaluated on three standard heterogeneous graph benchmarks, it consistently outperforms state-of-the-art HGNNs, achieving average accuracy gains of 3.2–5.8 percentage points. These results validate the effectiveness and generalizability of prompt-based alignment between pre-training objectives and downstream tasks.
📝 Abstract
Graphs have emerged as a natural choice to represent and analyze the intricate patterns and rich information of the Web, enabling applications such as online page classification and social recommendation. The prevailing ''pre-train, fine-tune'' paradigm has been widely adopted in graph machine learning tasks, particularly in scenarios with limited labeled nodes. However, this approach often exhibits a misalignment between the training objectives of pretext tasks and those of downstream tasks. This gap can result in the ''negative transfer'' problem, wherein the knowledge gained from pre-training adversely affects performance in the downstream tasks. The surge in prompt-based learning within Natural Language Processing (NLP) suggests the potential of adapting a ''pre-train, prompt'' paradigm to graphs as an alternative. However, existing graph prompting techniques are tailored to homogeneous graphs, neglecting the inherent heterogeneity of Web graphs. To bridge this gap, we propose HetGPT, a general post-training prompting framework to improve the predictive performance of pre-trained heterogeneous graph neural networks (HGNNs). The key is the design of a novel prompting function that integrates a virtual class prompt and a heterogeneous feature prompt, with the aim to reformulate downstream tasks to mirror pretext tasks. Moreover, HetGPT introduces a multi-view neighborhood aggregation mechanism, capturing the complex neighborhood structure in heterogeneous graphs. Extensive experiments on three benchmark datasets demonstrate HetGPT's capability to enhance the performance of state-of-the-art HGNNs on semi-supervised node classification.
Problem

Research questions and friction points this paper is trying to address.

Improves pre-trained heterogeneous graph neural networks.
Addresses misalignment in pretext and downstream tasks.
Introduces novel prompting for heterogeneous graph learning.
Innovation

Methods, ideas, or system contributions that make the work stand out.

Prompt tuning in heterogeneous graphs
Virtual and heterogeneous feature prompts
Multi-view neighborhood aggregation mechanism
Y
Yihong Ma
University of Notre Dame, Notre Dame, Indiana, USA
N
Ning Yan
Futurewei Technologies Inc., Santa Clara, California, USA
J
Jiayu Li
Syracuse University, Syracuse, New York, USA
M
Masood S. Mortazavi
Futurewei Technologies Inc., Santa Clara, California, USA
N
N. Chawla
University of Notre Dame, Notre Dame, Indiana, USA