基于跨模态生成模型的网络表征学习方法研究

项目摘要

网络表征学习是网络数据分析与建模的有效手段和重要研究方向，然而对于普遍存在的富含内容的网络来说，如何学习到可有效融合结构信息和内容语义的表征，仍是目前极具挑战的问题。对此，本项目创新性地将节点的上下文结构信息和丰富的内容信息统一地视作多模态信源，提出一整套跨模态网络表征学习方法。首先，针对结构与内容的融合问题，分别提出基于序列到序列及网络到网络的模态间映射变换的端到端多模态网络学习理论，一方面有效融合结构和内容信息，另一方面可自适应地学习高阶近邻关系。其次，为充分地从多模态信息中获取数据，本项目进一步提出基于对偶生成的多模态数据表征学习，通过双向生成任务获取更好的网络表征；再次，本项目进一步研究网络表征学习中监督信息的利用问题，基于强化学习引入弱监督学习等机制来学习更具区分性的网络表征。本项目研究成果是对网络表征学习、多模态学习等理论的探索和扩展，具有显著的理论和应用价值。

代表性成果

G-Prompt: Graphon-based Prompt Tuning for Graph Classification

Yutai Duan; Jie Liu*; Shaowei Chen; Liyi Chen; Jianhua Wu

Information Processing and Management, 2023

Abstract: Prompt tuning has been demonstrated to be effective in exploiting pre-trained models (PTMs) to perform downstream tasks. Motivated by this trend, researchers explored prompt tuning methods to leverage graph PTMs in downstream tasks related to graph classification. However, the inherent non-Euclidean and abstract characteristics of graph data present a set of challenging issues, which encompass the generation of graph-level prompts with accurate task-related knowledge, the enhancement of prompt adaptability to downstream tasks, and the prediction based on graph-level prompts. To address these issues, we propose Graphon-based Prompt Tuning (G-Prompt) as a systemic solution. G-Prompt consists of two elaborate modules. Specifically, we present a graph-level prompt generation (GP) module that describes the knowledge of downstream tasks by estimating graphons and generates graph-level prompts based on the knowledge. Subsequently, GP module also performs prompt ensembling steps and optimization based on downstream tasks to ensure the knowledge distribution of the prompts and their adaptability to downstream tasks. Furthermore, we propose a graph answer (GA) module to make accurate predictions based on multiple graph-level prompts. GA module models the conditional probability that the input graph belongs to the class corresponding to the prompt based on different prompts, respectively. It then selects the class with the highest probability as the prediction. Extensive experiments with 6 real-world datasets show that G-Prompt achieves state-of-the-art performance, outperforming compared methods by an average of 5%.

Contrastive Fine-tuning for Low-resource Graph-level Transfer Learning

Yutai Duan; Jie Liu*; Shaowei Chen; Jianhua Wu

lnformation Sciences, 2023.

Abstract: Due to insufficient supervision and the gap between pre-training pretext tasks and downstream tasks, transferring pre-trained graph neural networks (GNNs) to downstream tasks in low-resource scenarios remains challenging. In this paper, a Contrastive Fine-tuning (Con-tuning) framework is proposed for low-resource graph-level transfer learning, and a graph-level supervised contrastive learning (SCL) task is designed within the framework as the first attempt to introduce SCL for fine-tuning processes of pre-trained GNNs. The SCL task compensates for the insufficient supervision problem in low-resource scenarios and narrows the gap between pretext tasks and downstream tasks. To further reinforce the supervision signal in the SCL task, we devise a graphon theory based labeled graph generator to extract the generalized knowledge of a specific class of graphs. Based on this knowledge, graph-level templates are generated for each class and used as contrastive samples in the SCL task. Then, the proposed Con-tuning framework jointly learns the SCL task and downstream tasks to effectively fine-tune the pre-trained GNNs for downstream tasks. Extensive experiments with eight real-world datasets show that Con-tuning framework enables pre-trained GNNs to achieve better performance on graph-level downstream tasks in low-resource settings.

Leveraging maximum entropy and correlation on latent factors for learning representations

Zhicheng He; Jie Liu*; Kai Dang; Fuzhen Zhuang; Yalou Huang

Neural Networks, 2020, 2020(131)：312-323.

Abstract: Many tasks involve learning representations from matrices, and Non-negative Matrix Factorization (NMF) has been widely used due to its excellent interpretability. Through factorization, sample vectors are reconstructed as additive combinations of latent factors, which are represented as non-negative distributions over the raw input features. NMF models are significantly affected by latent factors’ distribution characteristics and the correlations among them. And NMF models are faced with the challenge of learning robust latent factor. To this end, we propose to learn representations with an awareness of the semantic quality evaluated from the aspects of intra- and inter-factors. On the one hand, a Maximum Entropy-based function is devised for the intra-factor semantic quality. On the other hand, the semantic uniqueness is evaluated via inter-factor correlation, which reinforces the aim of semantic compactness. Moreover, we present a novel non-linear NMF framework. The learning algorithm is presented and the convergence is theoretically analyzed and proved. Extensive experimental results on multiple datasets demonstrate that our method can be successfully applied to representative NMF models and boost performances over state-of-the-art models.

Network Embedding with Dual Generation Tasks

Na Li; Jie Liu*; Zhicheng He; Chunhai Zhang; Jiaying Xie

IEEE Transactions on Knowledge and Data Engineering, 2022.

Abstract: We study the problem of Network Embedding (NE) for content-rich networks. NE models aim to learn efficient low-dimensional dense vectors for network vertices which are crucial to many network analysis tasks. The core problem of content-rich network embedding is to learn and integrate the semantic information conveyed by network structure and node content. In this paper, we propose a general end-to-end model, Dual GEnerative Network Embedding (DGENE), to leverage the complementary information of network structure and content. In this model, each vertex is regarded as an object with two modalities: node identity and textual content. Then we formulate two dual generation tasks, Node Identification (NI) which recognizes nodes’ identities given their contents, and Content Generation (CG) which generates textual contents given the nodes’ identities. We develop specific Content2Node and Node2Content models for the two tasks. Under the DGENE framework, the two dual models are learned by sharing and integrating intermediate layers. Extensive experimental results show that our model yields a significant performance gain compared to the state-of-the-art NE methods. Moreover, our model has an interesting and useful byproduct, that is, a component of our model can generate texts and nodes, which is potentially useful for many tasks.

Multi-Scale Distillation from Multiple Graph Neural Networks

Chunhai Zhang; Jie Liu*; Kai Dang; Wenzheng Zhang

Association for the Advancement of Artificial Intelligence (AAAI), 2022, 36(4)：4337.

Abstract: Knowledge Distillation (KD), which is an effective model compression and acceleration technique, has been successfully applied to graph neural networks (GNNs) recently. Existing approaches utilize a single GNN model as the teacher to distill knowledge. However, we notice that GNN models with different number of layers demonstrate different classification abilities on nodes with different degrees. On the one hand, for nodes with high degrees, their local structures are dense and complex, hence more message passing is needed. Therefore, GNN models with more layers perform better. On the other hand, for nodes with low degrees, whose local structures are relatively sparse and simple, the repeated message passing can easily lead to over-smoothing. Thus, GNN models with less layers are more suitable. However, existing single-teacher GNN knowledge distillation approaches which are based on a single GNN model, are sub-optimal. To this end, we propose a novel approach to distill multi-scale knowledge, which learns from multiple GNN teacher models with different number of layers to capture the topological semantic at different scales. Instead of learning from the teacher models equally, the proposed method automatically assigns proper weights for each teacher model via an attention mechanism which enables the student to select teachers for different local structures. Extensive experiments are conducted to evaluate the proposed method on four public datasets. The experimental results demonstrate the superiority of our proposed method over state-of-the-art methods. Our code is publicly available at https://github.com/NKU-IIPLab/MSKD.

DASH: An Agile Knowledge Graph System Disentangling Demands, Algorithms, Data Resources, and Humans

Shaowei Chen; Haoran Wang; Jie Liu*; Jiahui Wu

CIKM, Atlanta, GA, USA, 2022-10-17至2022-10-21.

Abstract: Knowledge graph (KG) is an important branch of artificial intelligence, which has attracted increasing research interest. However, in most enterprises, it is challenging to quickly construct KGs with multi-source and heterogeneous data and apply KGs to meet diverse business demands. To deal with these challenges, we propose an agile knowledge graph system following the novel principle of disentangling Demands, Algorithms, data reSources, and Humans (DASH). Specifically, our system is equipped with prior information-based knowledge extraction, self-supervised knowledge integration, and hierarchical knowledge base question answering algorithms that have outstanding generalizability and portability. Meanwhile, we propose a semi-automatic data accumulation framework to reduce labor costs of data annotations. Based on DASH, we develop a Web application with easy-to-use functionalities such as canvases and drag-and-drop, and illustrate its usage in a financial scenario.