Abstract:Active learning reduces the annotation cost of machine learning by selecting and querying informative unlabeled samples. Semi-supervised active learning methods can considerably utilize the regional information of unlabeled samples, and thus, more effectively select valuable samples. Existing semi-supervised batch active learning algorithms frequently exhibit poor robustness due to their high computational complexity, making handling large-scale datasets a difficult task. However, existing active learning algorithms based on high-performance semi-supervised learners adopt a single-sample selection mode, under which the model requires multiple rounds of iterative processes, significantly reducing the overall efficiency of the algorithm and affecting its practicality. To address these issues, we propose a new semi-supervised batch active learning algorithm called approximate error reduction based on mutual information (MIAER). First, we use hierarchical anchor graph regularization (HAGR) as the semi-supervised learner. HAGR exhibits good robustness and only involves a small-scale reduced Laplacian matrix in its optimization process, enabling rapid processing of large-scale datasets. Second, we propose a batch sampling strategy based on mutual information and error reduction in the sample selection stage. This strategy, which is based on hierarchical anchor graphs, first measures the uncertainty of samples by using approximate error reduction, considerably reducing computational overhead. Then, it uses mutual information to measure the diversity of samples in category space while removing redundant batch samples, preserving samples with high uncertainty as much as possible. Comparative experiments with several advanced active learning methods on a large number of datasets fully demonstrate the effectiveness and stability of MIAER.

Non-myopic active learning with mutual information

A near-optimal non-myopic active learning method

Non-myopic Active Learning with Performance Guarantee

Batch Mode Active Learning and Its Application to Medical Image Classification

Semi-supervised batch active learning based on mutual information

Cost-effective Batch-mode Multi-label Active Learning

Semisupervised SVM Batch Mode Active Learning with Applications to Image Retrieval

Multi-class Active Learning: A Hybrid Informative and Representative Criterion Inspired Approach.

Batch-Mode Active Learning via Error Bound Minimization.

Active Learning with Nonparallel Support Vector Machine for Binary Classification

Semi-supervised SVM Batch Mode Active Learning for Image Retrieval

Multi-label Active Learning Based on Submodular Functions

A Multicriterion Query-Based Batch Mode Active Learning Technique

Bayesian Active Learning by Soft Mean Objective Cost of Uncertainty

Multiple-view multiple-learner active learning

Exploring Representativeness and Informativeness for Active Learning.

A Method of Active Learning with Optimal Sampling Strategy

Unlabeled data selection for active learning in image classification

Efficient Active Learning by Querying Discriminative and Representative Samples and Fully Exploiting Unlabeled Data

K-means Based on Active Learning for Support Vector Machine.

A novel batch-mode active learning method for SVM classifier