[论文解读] Sampling-guided exploration of active feature selection policies

Gabriel Bernardino, Anders Jönsson+2 more|arXiv (Cornell University)|Mar 16, 2026

Machine Learning and Data Classification被引用 0

一句话总结

论文在更大数据集上扩展了基于强化学习的实例特定主动特征选择框架，采用基于启发式的策略探索和后拟合正则化来产生紧凑、可解释的获取策略，同时提高准确性并降低策略复杂性。

ABSTRACT

Determining the most appropriate features for machine learning predictive models is challenging regarding performance and feature acquisition costs. In particular, global feature choice is limited given that some features will only benefit a subset of instances. In previous work, we proposed a reinforcement learning approach to sequentially recommend which modality to acquire next to reach the best information/cost ratio, based on the instance-specific information already acquired. We formulated the problem as a Markov Decision Process where the state's dimensionality changes during the episode, avoiding data imputation, contrary to existing works. However, this only allowed processing a small number of features, as all possible combinations of features were considered. Here, we address these limitations with two contributions: 1) we expand our framework to larger datasets with a heuristic-based strategy that focuses on the most promising feature combinations, and 2) we introduce a post-fit regularisation strategy that reduces the number of different feature combinations, leading to compact sequences of decisions. We tested our method on four binary classification datasets (one involving high-dimensional variables), the largest of which had 56 features and 4500 samples. We obtained better performance than state-of-the-art methods, both in terms of accuracy and policy complexity.

研究动机与目标

• 在获取成本下按实例选择信息性特征的挑战进行动机阐述并提出解决方案。
• 提出一个可扩展的基于RL的框架，通过使用超状态（不连通的特征子集）和DAG结构来避免填充缺失数据。
• 引入一个基于采样的启发式方法，聚焦于大规模特征集合中有前景的特征组合以进行策略学习。
• 提出一种正则化策略，减少不同获取序列的数量并提升可解释性。

提出的方法

• 将主动特征选择建模为在超状态上的马尔可夫决策过程，允许在不插补缺失数据的情况下进行按实例的特征获取决策。
• 通过为每个超状态训练一个独立的回归器来独立估计Q(s,a)，扩展基于Bellman的Q值计算。
• 使用图探索启发式在有前景的超状态子集上构建受限策略，减少估计量的指数增长。
• 加入稀疏策略正则化（受REPS启发），增加一个虚拟奖励项以惩罚访问太多的超状态，从而提升可解释性并减少过拟合。

实验结果

研究问题

• RQ1 如何在不枚举所有特征组合的情况下将主动特征选择扩展到中/大规模特征集？
• RQ2 基于采样的探索策略是否能够识别对实例决策最有价值的特征组合？
• RQ3 正则化是否能够在不牺牲准确性的前提下降低主动特征选择的策略复杂性？
• RQ4 与最先进的主动特征选择方法在准确性和成本效率方面在多样数据集上的比较如何？

主要发现

• 所提出的基于采样引导的探索在多个数据集上实现了与EDDI和Wang的聚类相比的竞争性或更优的准确性，包括一个高维案例。
• 该方法能够处理高达56个特征和4500个样本的数据集，策略复杂性（不同获取序列数量更少）和可解释性得到提升。
• 在Heart数据集上的实际成本评估显示，当成本变化时，该方法选择更便宜、信息量更大的特征，而不是在同质成本设置下倾向于昂贵特征。
• 基于启发式的策略扩展在大规模特征集（如Spam）上加速探索，在准确性与成本之间实现更好的权衡，优于随机探索。