Dynamic Feature Selection on Variable Feature Sets using Features of Features

Machine learning models usually assume that the value of input features are revealed in advance. However, in many real-world problems, a cost is associated with measuring these features. To reduce measurement costs, various methods have been proposed to dynamically select which features to measure, but existing methods assume that the set of measurable features remains fixed, which makes them unsuitable for cases where the set of measurable features varies from instance to instance. To overcome this limitation, we define a novel problem setting for Dynamic Feature Selection (DFS) with variable feature sets and propose a deep learning method that utilizes prior information about each feature, referred to as "features of features". Experimental results on several datasets demonstrate that the proposed method effectively selects features based on the prior information, even when measurable features set changes from instance to instance.