Stochastic Encodings for Active Feature Acquisition
Abstract
Active Feature Acquisition is an instance-wise, sequential decision making problem. The aim is to dynamically select which feature to measure based on current observations, independently for each test instance. Common approaches either use Reinforcement Learning, which experiences training difficulties, or greedily maximize the conditional mutual information of the label and unobserved features, which makes myopic acquisitions. To address these shortcomings, we introduce a latent variable model, trained in a supervised manner. Acquisitions are made by reasoning about the features across many possible unobserved realizations in a stochastic latent space. Extensive evaluation on a large range of synthetic and real datasets demonstrates that our approach reliably outperforms a diverse set of baselines.
Cite
Text
Norcliffe et al. "Stochastic Encodings for Active Feature Acquisition." Proceedings of the 42nd International Conference on Machine Learning, 2025.Markdown
[Norcliffe et al. "Stochastic Encodings for Active Feature Acquisition." Proceedings of the 42nd International Conference on Machine Learning, 2025.](https://mlanthology.org/icml/2025/norcliffe2025icml-stochastic/)BibTeX
@inproceedings{norcliffe2025icml-stochastic,
title = {{Stochastic Encodings for Active Feature Acquisition}},
author = {Norcliffe, Alexander Luke Ian and Lee, Changhee and Imrie, Fergus and Van Der Schaar, Mihaela and Lio, Pietro},
booktitle = {Proceedings of the 42nd International Conference on Machine Learning},
year = {2025},
pages = {46784-46814},
volume = {267},
url = {https://mlanthology.org/icml/2025/norcliffe2025icml-stochastic/}
}