Interactive reinforced feature selection with traverse strategy

• Published in: Knowledge and information systems 2023-05, Vol.65 (5), p.1935-1962
• Main Authors: Liu, Kunpeng, Wang, Dongjie, Du, Wan, Wu, Dapeng Oliver, Fu, Yanjie
• Format: Article
• Language: English
• Subjects: Cognitive tasks; Computational efficiency; Computer Science; Computing costs; Data Mining and Knowledge Discovery; Database Management; Efficiency; Feature selection; Importance sampling; Information Storage and Retrieval; Information Systems and Communication Service; Information Systems Applications (incl.Internet); IT in Business; Machine learning; Multiagent systems; Regular Paper; Training
• ISSN: 0219-1377; 0219-3116
• DOI: 10.1007/s10115-022-01812-3

In this paper, we propose a single-agent Monte Carlo-based reinforced feature selection method, as well as two efficiency improvement strategies, i.e., early stopping strategy and reward-level interactive strategy. Feature selection is one of the most important technologies in data prepossessing, aiming to find the optimal feature subset for a given downstream machine learning task. Enormous research has been done to improve its effectiveness and efficiency. Recently, the multi-agent reinforced feature selection (MARFS) has achieved great success in improving the performance of feature selection. However, MARFS suffers from the heavy burden of computational cost, which greatly limits its application in real-world scenarios. In this paper, we propose an efficient reinforcement feature selection method, which uses one agent to traverse the whole feature set and decides to select or not select each feature one by one. Specifically, we first develop one behavior policy and use it to traverse the feature set and generate training data. And then, we evaluate the target policy based on the training data and improve the target policy by Bellman equation. Besides, we conduct the importance sampling in an incremental way and propose an early stopping strategy to improve the training efficiency by the removal of skew data. In the early stopping strategy, the behavior policy stops traversing with a probability inversely proportional to the importance sampling weight. In addition, we propose a reward-level and training-level interactive strategy to improve the training efficiency via external advice. What’s more, we propose an incremental descriptive statistics method to represent the state with low computational cost. Finally, we design extensive experiments on real-world data to demonstrate the superiority of the proposed method.