Main content area

Greedy active learning algorithm for logistic regression models

Hsu, Hsiang-Ling, Chang, Yuan-chin Ivan, Chen, Ray-Bing
Computational statistics & data analysis 2019 v.129 pp. 119-134
algorithms, data collection, experimental design, models, regression analysis, telescopes
We study a logistic model-based active learning procedure for binary classification problems, in which we adopt a batch subject selection strategy with a modified sequential experimental design method. Moreover, accompanying the proposed subject selection scheme, we simultaneously conduct a greedy variable selection procedure such that we can update the classification model with all labeled training subjects. The proposed algorithm repeatedly performs both subject and variable selection steps until a prefixed stopping criterion is reached. Our numerical results show that the proposed procedure has competitive performance, with smaller training size and a more compact model compared with that of the classifier trained with all variables and a full data set. We also apply the proposed procedure to a well-known wave data set (Breiman et al., 1984) and a MAGIC gamma telescope data set to confirm the performance of our method.