paper-with-me

Papers

Graph Policy Network for Transferable Active Learning on Graphs

2020-06-24 · NeurIPS 2020 12 · Shengding Hu, Zheng Xiong, Meng Qu, Xingdi Yuan, Marc-Alexandre Côté, Zhiyuan Liu, Jian Tang

Graph neural networks (GNNs) have been attracting increasing popularity due to their simplicity and effectiveness in a variety of fields. However, a large number of labeled data is generally required to train these networks, which could be very expensive to obtain in some domains. In this paper, we study active learning for GNNs, i.e., how to efficiently label the nodes on a graph to reduce the annotation cost of training GNNs. We formulate the problem as a sequential decision process on graphs and train a GNN-based policy network with reinforcement learning to learn the optimal query strategy. By jointly training on several source graphs with full labels, we learn a transferable active learning policy which can directly generalize to unlabeled target graphs. Experimental results on multiple datasets from different domains prove the effectiveness of the learned policy in promoting active learning performance in both settings of transferring between graphs in the same domain and across different domains.

📄 PDF Abstract BibTeX arXiv:2006.13463

Code (1)

ShengdingHu/GraphPolicyNetworkActiveLearning 공식 구현 pytorch

Tasks

Active Learning

Similar Papers 제목 키워드 기반

Transfer Active Learning For Graph Neural Networks

2019-09-25 · Shengding Hu, Meng Qu, Zhiyuan Liu, Jian Tang

Graph neural networks have been proved very effective for a variety of prediction tasks on graphs such as node classification. Generally, a large number of labeled data are required to train these networks. However, in r…

Active LearningNode ClassificationTransfer Learning

GraphMimic: Graph-to-Graphs Generative Modeling from Videos for Policy Learning

2025-01-01 · CVPR 2025 1 · Guangyan Chen, Te Cui, Meiling Wang, Chengcai Yang 외

Learning from demonstration is a powerful method for robotic skill acquisition. However, the significant expense of collecting such action-labeled robot data presents a major bottleneck. Video data, a rich data sourc…

Graph Generation

Decentralized Learning Strategies for Estimation Error Minimization with Graph Neural Networks

2026-01-19 · Xingran Chen, Navid NaderiAlizadeh, Alejandro Ribeiro, Shirin Saeedi Bidokhti arxiv

We address real-time sampling and estimation of autoregressive Markovian sources in dynamic yet structurally similar multi-hop wireless networks. Each node caches samples from others and communicates over wireless collis…

Multi-agent Reinforcement Learning

Graph-Structured Gaussian Processes for Transferable Graph Learning

2023-09-21 · NeurIPS 2023 11

Transferable graph learning involves knowledge transferability from a source graph to a relevant target graph. The major challenge of transferable graph learning is the distribution shift between source and target graphs…

Learning Transferable Skills in Action RPGs via Directed Skill Graphs and Selective Adaptation

2026-01-25 · Ali Najar arxiv

Lifelong agents should expand their competence over time without retraining from scratch or overwriting previously learned behaviors. We investigate this in a challenging real-time control setting (Dark Souls III) by rep…