Model-Change Active Learning in Graph-Based Semi-Supervised Learning
This work addresses the problem of efficient label acquisition in semi-supervised learning for researchers and practitioners, but it is incremental as it builds on existing active learning and graph-based methods.
The paper tackles the challenge of selecting which unlabeled points to label in graph-based semi-supervised learning to improve classifier accuracy while limiting label costs, by introducing a 'Model Change' active learning method that quantifies classifier changes from new labels and shows improved performance over prior state-of-the-art in multiclass examples.
Active learning in semi-supervised classification involves introducing additional labels for unlabelled data to improve the accuracy of the underlying classifier. A challenge is to identify which points to label to best improve performance while limiting the number of new labels. "Model Change" active learning quantifies the resulting change incurred in the classifier by introducing the additional label(s). We pair this idea with graph-based semi-supervised learning methods, that use the spectrum of the graph Laplacian matrix, which can be truncated to avoid prohibitively large computational and storage costs. We consider a family of convex loss functions for which the acquisition function can be efficiently approximated using the Laplace approximation of the posterior distribution. We show a variety of multiclass examples that illustrate improved performance over prior state-of-art.