Active Sampling for Node Attribute Completion on Graphs

TL;DR

This paper introduces ATS, an active sampling algorithm that effectively restores missing node attributes in graphs by measuring node importance through structure and uncertainty, outperforming existing methods.

Contribution

It proposes a novel active sampling method that combines representativeness and uncertainty for node attribute completion, addressing limitations of prior decoupled frameworks.

Findings

ATS outperforms baseline methods on benchmark datasets.

The weighting scheme effectively balances structure and uncertainty.

Experiments demonstrate improved accuracy in downstream tasks.

Abstract

Node attribute, a type of crucial information for graph analysis, may be partially or completely missing for certain nodes in real world applications. Restoring the missing attributes is expected to benefit downstream graph learning. Few attempts have been made on node attribute completion, but a novel framework called Structure-attribute Transformer (SAT) was recently proposed by using a decoupled scheme to leverage structures and attributes. SAT ignores the differences in contributing to the learning schedule and finding a practical way to model the different importance of nodes with observed attributes is challenging. This paper proposes a novel AcTive Sampling algorithm (ATS) to restore missing node attributes. The representativeness and uncertainty of each node's information are first measured based on graph structure, representation similarity and learning bias. To select nodes as train samples in the next optimization step, a weighting scheme controlled by Beta distribution is then introduced to linearly combine the two properties. Extensive experiments on four public benchmark datasets and two downstream tasks have shown the superiority of ATS in node attribute completion.