Bridged-GNN: Knowledge Bridge Learning for Effective Knowledge Transfer
arXiv:2308.09499 · doi:10.1145/3583780.3614796
Abstract
The data-hungry problem, characterized by insufficiency and low-quality of data, poses obstacles for deep learning models. Transfer learning has been a feasible way to transfer knowledge from high-quality external data of source domains to limited data of target domains, which follows a domain-level knowledge transfer to learn a shared posterior distribution. However, they are usually built on strong assumptions, e.g., the domain invariant posterior distribution, which is usually unsatisfied and may introduce noises, resulting in poor generalization ability on target domains. Inspired by Graph Neural Networks (GNNs) that aggregate information from neighboring nodes, we redefine the paradigm as learning a knowledge-enhanced posterior distribution for target domains, namely Knowledge Bridge Learning (KBL). KBL first learns the scope of knowledge transfer by constructing a Bridged-Graph that connects knowledgeable samples to each target sample and then performs sample-wise knowledge transfer via GNNs.KBL is free from strong assumptions and is robust to noises in the source data. Guided by KBL, we propose the Bridged-GNN} including an Adaptive Knowledge Retrieval module to build Bridged-Graph and a Graph Knowledge Transfer module. Comprehensive experiments on both un-relational and relational data-hungry scenarios demonstrate the significant improvements of Bridged-GNN compared with SOTA methods
Accepted by CIKM2023
References in corpus (33)
- Learning Transferable Features with Deep Adaptation Networks
- Unsupervised Domain Adaptation by Backpropagation
- Conditional Adversarial Domain Adaptation
- Simple and Deep Graph Convolutional Networks
- Contrastive Cross-Domain Sequential Recommendation
- Learning Multi-granularity User Intent Unit for Session-based Recommendation
- Discovering Invariant Rationales for Graph Neural Networks
- TIMME: Twitter Ideology-detection via Multi-task Multi-relational Embedding
- A Survey of Knowledge Graph Reasoning on Graph Types: Static, Dynamic, and Multimodal
- Company-as-Tribe: Company Financial Risk Assessment on Tribe-Style Graph with Hierarchical Graph Neural Networks
- Higher-order Clustering and Pooling for Graph Neural Networks
- Homophily-oriented Heterogeneous Graph Rewiring
- Distinguishing manipulated stocks via trading network analysis
- Learning on Graphs with Out-of-Distribution Nodes
- Handling Distribution Shifts on Graphs: An Invariance Perspective
- GOOD: A Graph Out-of-Distribution Benchmark
- MM-GNN: Mix-Moment Graph Neural Network towards Modeling Neighborhood Feature Distribution
- Cross-Domain Aspect Extraction using Transformers Augmented with Knowledge Graphs
- GraphGLOW: Universal and Generalizable Structure Learning for Graph Neural Networks
- Predicting the Silent Majority on Graphs: Knowledge Transferable Graph Neural Network
- Beyond Homophily: Reconstructing Structure for Graph-agnostic Clustering
- A Simple Baseline to Semi-Supervised Domain Adaptation for Machine Translation
- Contrastive Domain Adaptation for Early Misinformation Detection: A Case Study on COVID-19
- Weak-shot Fine-grained Classification via Similarity Transfer
- Deconfounding to Explanation Evaluation in Graph Neural Networks
- From Known to Unknown: Quality-aware Self-improving Graph Neural Network for Open Set Social Event Detection
- Make Heterophily Graphs Better Fit GNN: A Graph Rewiring Approach
- Distributionally Robust Semi-Supervised Learning Over Graphs
- Initialization for Network Embedding: A Graph Partition Approach
- Company Competition Graph
- Node Embedding from Neural Hamiltonian Orbits in Graph Neural Networks
- Diffusion-Jump GNNs: Homophiliation via Learnable Metric Filters
- Inclusive FinTech Lending via Contrastive Learning and Domain Adaptation