Uncertainty for Active Learning on Graphs

arXiv:2405.01462v1 Announce Type: new
Abstract: Uncertainty Sampling is an Active Learning strategy that aims to improve the data efficiency of machine learning models by iteratively acquiring labels of data points with the highest uncertainty. While it has proven effective for independent data its applicability to graphs remains under-explored. We propose the first extensive study of Uncertainty Sampling for node classification: (1) We benchmark Uncertainty Sampling beyond predictive uncertainty and highlight a significant performance gap to other Active Learning strategies. (2) …

abstract active learning arxiv cs.lg data efficiency graphs independent labels machine machine learning machine learning models sampling strategy study type uncertainty while