An official website of the United States government
Graph neural networks (GNNs) have shown remarkable performance on diverse graph mining tasks. While sharing the same message passing framework, our study shows that different GNNs learn distinct knowledge from the same graph. This implies potential performance improvement by distilling the complementary knowledge from multiple models. However, knowledge distillation (KD) transfers knowledge from high-capacity teachers to a lightweight student, which deviates from our scenario: GNNs are often shallow. To transfer knowledge effectively, we need to tackle two challenges: how to transfer knowledge from compact teachers to a student with the same capacity; and, how to exploit student GNN's own learning ability. In this paper, we propose a novel adaptive KD framework, called BGNN, which sequentially transfers knowledge from multiple GNNs into a student GNN. We also introduce an adaptive temperature module and a weight boosting module. These modules guide the student to the appropriate knowledge for effective learning. Extensive experiments have demonstrated the effectiveness of BGNN. In particular, we achieve up to 3.05% improvement for node classification and 6.35% improvement for graph classification over vanilla GNNs.
more »
« less
This content will become publicly available on March 17, 2027
Pareto-Based Heterogeneous Knowledge Distillation for MLPs on Graphs
Heterogeneous Graph Neural Networks (HGNNs) have demonstrated remarkable capabilities in capturing effective information in heterogeneous graphs, achieving outstanding performance in various learning tasks. However, the heavy dependency of HGNNs on neighbors information may result in high latency, which restricts their practicality in real-world applications. Recent studies have attempted to overcome such latency in Graph Neural Networks (GNNs) by distilling knowledge into student models that do not rely on graph structure. But these approaches primarily focus on replicating teachers' predictive outcomes while neglecting the structural knowledge they encoded. This limitation makes such approach less effective when graphs become complex, particularly on heterogeneous graphs. Motivated by this challenge, we propose HGKD, a novel hierarchical knowledge distillation framework that transfers both structural knowledge and predictive outcomes from HGNN teachers to a multi-layer perceptron student. Additionally, we provide two variants of HGKD that help the student learn from multiple teacher models through Pareto learning and incorporate low-cost neighbor information. We evaluate HGKD and its variants on a range of heterogeneous graph datasets. The results demonstrate that our student model achieves performance comparable to or exceeding that of HGNN teachers, despite not relying on graph structures during inference.
more »
« less
- Award ID(s):
- 2528540
- PAR ID:
- 10683249
- Publisher / Repository:
- Proceedings of the AAAI Conference on Artificial Intelligence (AAAI)
- Date Published:
- Journal Name:
- Proceedings of the AAAI Conference on Artificial Intelligence
- Volume:
- 40
- Issue:
- 34
- ISSN:
- 2159-5399
- Page Range / eLocation ID:
- 28751 to 28759
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract BackgroundThe recent development of high-throughput sequencing has created a large collection of multi-omics data, which enables researchers to better investigate cancer molecular profiles and cancer taxonomy based on molecular subtypes. Integrating multi-omics data has been proven to be effective for building more precise classification models. Most current multi-omics integrative models use either an early fusion in the form of concatenation or late fusion with a separate feature extractor for each omic, which are mainly based on deep neural networks. Due to the nature of biological systems, graphs are a better structural representation of bio-medical data. Although few graph neural network (GNN) based multi-omics integrative methods have been proposed, they suffer from three common disadvantages. One is most of them use only one type of connection, either inter-omics or intra-omic connection; second, they only consider one kind of GNN layer, either graph convolution network (GCN) or graph attention network (GAT); and third, most of these methods have not been tested on a more complex classification task, such as cancer molecular subtypes. ResultsIn this study, we propose a novel end-to-end multi-omics GNN framework for accurate and robust cancer subtype classification. The proposed model utilizes multi-omics data in the form of heterogeneous multi-layer graphs, which combine both inter-omics and intra-omic connections from established biological knowledge. The proposed model incorporates learned graph features and global genome features for accurate classification. We tested the proposed model on the Cancer Genome Atlas (TCGA) Pan-cancer dataset and TCGA breast invasive carcinoma (BRCA) dataset for molecular subtype and cancer subtype classification, respectively. The proposed model shows superior performance compared to four current state-of-the-art baseline models in terms of accuracy, F1 score, precision, and recall. The comparative analysis of GAT-based models and GCN-based models reveals that GAT-based models are preferred for smaller graphs with less information and GCN-based models are preferred for larger graphs with extra information.more » « less
-
Recent years have witnessed the superior performance of heterogeneous graph neural networks (HGNNs) in dealing with heterogeneous information networks (HINs). Nonetheless, the success of HGNNs often depends on the availability of sufficient labeled training data, which can be very expensive to obtain in real scenarios. Active learning provides an effective solution to tackle the data scarcity challenge. For the vast majority of the existing work regarding active learning on graphs, they mainly focus on homogeneous graphs, and thus fall in short or even become inapplicable on HINs. In this paper, we study the active learning problem with HGNNs and propose a novel meta-reinforced active learning framework MetRA. Previous reinforced active learning algorithms train the policy network on labeled source graphs and directly transfer the policy to the target graph without any adaptation. To better exploit the information from the target graph in the adaptation phase, we propose a novel policy transfer algorithm based on meta-Q-learning termed per-step MQL. Empirical evaluations on HINs demonstrate the effectiveness of our proposed framework. The improvement over the best baseline is up to 7% in Micro-F1.more » « less
-
A prevalent approach of entity-oriented systems involves retrieving relevant entities by harnessing knowledge graph embeddings. These embeddings encode entity information in the context of the knowledge graph and are static in nature. Our goal is to generate entity embeddings that capture what renders them relevant for the query. This differs from entity embeddings constructed with static resource, for example, E-BERT. Previously, ~\citet{dalton2014entity} demonstrated the benefits obtained with the Entity Context Model, a pseudo-relevance feedback approach based on entity links in relevant contexts. In this work, we reinvent the Entity Context Model (ECM) for neural graph networks and incorporate pre-trained embeddings. We introduce three entity ranking models based on fundamental principles of ECM: (1) \acl{GAN}, (2) Simple Graph Relevance Networks, and (3) Graph Relevance Networks. \acl{GAN} and Graph Relevance Networks are the graph neural variants of ECM, that employ attention mechanism and relevance information of the relevant context respectively to ascertain entity relevance. Our experiments demonstrate that our neural variants of the ECM model significantly outperform the state-of-the-art BERT-ER ~\cite{10.1145/3477495.3531944} by more than 14\% and exceeds the performance of systems that use knowledge graph embeddings by over 101\%. Notably, our findings reveal that leveraging the relevance of the relevant context is more effective at identifying relevant entities than the attention mechanism. To evaluate the efficacy of the models, we conduct experiments on two standard benchmark datasets, DBpediaV2 and TREC Complex Answer Retrieval. To aid reproducibility, our code and data are available. https://github.com/TREMA-UNH/neural-entity-context-modelsmore » « less
-
This work considers the task of representation learning on the attributed relational graph (ARG). Both the nodes and edges in an ARG are associated with attributes/features allowing ARGs to encode rich structural information widely observed in real applications. Existing graph neural networks offer limited ability to capture complex interactions within local structural contexts, which hinders them from taking advantage of the expression power of ARGs. We propose motif convolution module (MCM), a new motif-based graph representation learning technique to better utilize local structural information. The ability to handle continuous edge and node features is one of MCM’s advantages over existing motif-based models. MCM builds a motif vocabulary in an unsupervised way and deploys a novel motif convolution operation to extract the local structural context of individual nodes, which is then used to learn higher level node representations via multilayer perceptron and/or message passing in graph neural networks. When compared with other graph learning approaches to classifying synthetic graphs, our approach is substantially better at capturing structural context. We also demonstrate the performance and explainability advantages of our approach by applying it to several molecular benchmarks.more » « less
