2021年3月8日 · In this paper, we propose a novel graph neural network framework, long-short distance aggregation networks (LSDAN), to overcome these limitations. By generating multiple graphs at different distance levels, based on the adjacency matrix, we develop a long-short distance attention model to model these graphs.

2024年12月1日 · Graph positive-unlabeled (GPU) learning aims to learn binary classifiers from only positive and unlabeled (PU) nodes. The state-of-the-art methods rely on provided class prior probabilistic and their performance lags far behind the fully labeled counterparts.

In this work, we propose GRAB (Graph-based Risk minimization with iterAtive Belief propagation), a novel end-to-end approach for graph-based PU learning that requires no class prior. GRAB models a given graph as a Markov network and runs …

2025年2月3日 · To deal with these issues, this paper proposes a novel model-agnostic PU learning algorithm for GNN-based link prediction by means of Positive-Unlabeled Area Under the Receiver Operating Characteristic Curve (PU-AUC) optimization. The proposed method is free of class prior estimation and able to handle the data imbalance.

2022年10月31日 · Based on the graph representation, PUTraceAD trains a GNN- and PU learning-based trace anomaly detection model. During the process, PU (Positive and Unlabeled) learning optimizes model parameters through estimating the data distribution.

Specifically, we propose the positive-unlabeled GNN (PU-GNN) with a distance-aware PU loss (section 3.1) as well as a regularizer from graph structure (section 3.2) to help model training.

2022年6月30日 · We compare GRAB with previous models for graph-based PU learning, including those for unsupervised representation learning. We adopt a graph convolutional network (GCN) as a base classifier for GRAB and most baselines.

In this article, we propose a novel GNN framework, long-short distance aggregation networks, to overcome these limitations. By generating multiple graphs at different distance levels, based on the adjacency matrix, we develop a long-short distance attention model to model these graphs.

The proposed PU-GNN achieves the best performance for almost all data sets and label ratios, demonstrating that graph structure plays a critical role in the loss design for PU node classification.

PUTraceAD: Trace Anomaly Detection with Partial Labels based on GNN and PU Learning Resources