Stochastic Encodings for Active Feature Acquisition
This addresses the challenge of efficient feature measurement in sequential decision-making, though it appears incremental as it builds on existing methods like reinforcement learning and mutual information maximization.
The paper tackles the problem of active feature acquisition, where features are selected sequentially for each test instance, by introducing a latent variable model that reasons about unobserved features in stochastic latent space. The approach outperforms diverse baselines on synthetic and real datasets.
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.