Generative Flow Networks for Precise Reward-Oriented Active Learning on Graphs
This addresses the challenge of limited exploration and reward proportionality in graph active learning for applications like graph neural networks, representing a novel method rather than an incremental improvement.
The paper tackles the problem of graph active learning by formulating it as a generative process with GFlowGNN, which generates samples proportional to rewards, resulting in improved exploration and transferability that outperforms state-of-the-art methods on real datasets.
Many score-based active learning methods have been successfully applied to graph-structured data, aiming to reduce the number of labels and achieve better performance of graph neural networks based on predefined score functions. However, these algorithms struggle to learn policy distributions that are proportional to rewards and have limited exploration capabilities. In this paper, we innovatively formulate the graph active learning problem as a generative process, named GFlowGNN, which generates various samples through sequential actions with probabilities precisely proportional to a predefined reward function. Furthermore, we propose the concept of flow nodes and flow features to efficiently model graphs as flows based on generative flow networks, where the policy network is trained with specially designed rewards. Extensive experiments on real datasets show that the proposed approach has good exploration capability and transferability, outperforming various state-of-the-art methods.