Learning with Feature Evolvable Streams
This addresses a practical issue in streaming data applications where feature evolution occurs, offering a solution for domains like sensor networks, but it is incremental as it builds on existing streaming learning methods.
The paper tackles the problem of learning from data streams where features evolve over time, such as when sensors are replaced, by proposing a novel paradigm called Feature Evolvable Streaming Learning that recovers vanished features to improve performance. Experiments on synthetic and real data validate the effectiveness, showing improved performance with theoretical guarantees.
Learning with streaming data has attracted much attention during the past few years. Though most studies consider data stream with fixed features, in real practice the features may be evolvable. For example, features of data gathered by limited-lifespan sensors will change when these sensors are substituted by new ones. In this paper, we propose a novel learning paradigm: \emph{Feature Evolvable Streaming Learning} where old features would vanish and new features would occur. Rather than relying on only the current features, we attempt to recover the vanished features and exploit it to improve performance. Specifically, we learn two models from the recovered features and the current features, respectively. To benefit from the recovered features, we develop two ensemble methods. In the first method, we combine the predictions from two models and theoretically show that with the assistance of old features, the performance on new features can be improved. In the second approach, we dynamically select the best single prediction and establish a better performance guarantee when the best model switches. Experiments on both synthetic and real data validate the effectiveness of our proposal.