面向低成本时序预测的主动特征获取：用树蒸馏学习可解释的 LAFA 策略
Active Feature Acquisition for Cost-Efficient Temporal Prediction with Reduced Participant Burden
针对纵向主动特征获取（LAFA）现有方法多依赖难以解释的神经网络，研究者提出一种树蒸馏方法，从 NN 版 LAFA 网络中学习可解释的选取策略。该方法在模拟实验和预测每日饮酒量的真实 EMA 数据集上验证，均能在精度损失极小的前提下显著减少每次采集的条目数。
Authors:Yunni Qu (1), Bing Cai Kok (2 and 3), Whitney Ringwald (4), Grant King (5), Aidan Wright (5), Kathleen Gates (2), Junier Oliva (1) ((1) Department of Computer Science, University of North Carolina at Chapel Hill, (2) Department of Psychology and Neuroscience, University of North Carolina at Chapel Hill, (3) School of Social Sciences, Nanyang Technological University, Singapore, (4) Department of Psychology, University of Minnesota Twin Cities, (5) Department of Psychology, University of Michigan)
Abstract:Accurate forecasting of pathological outcomes is a central problem in psychology. To do so, psychologists often collect intensive longitudinal data. However, in such studies, the desire to acquire a large number of variables for the sake of accurate prediction is often counteracted by the need to minimize participant burden. Acquiring more variables per occasion can yield better predictions, but having too many acquisitions increase the risk of non-response and attrition. Longitudinal Active Feature Acquisition (LAFA) is a principled approach to resolve this conundrum. Instead of requiring responses to every item at every acquisition occasion, LAFA produces a policy that seeks to optimally select dynamic subsets of items to be acquired at each timepoint while preserving our ability to forecast a specific outcome. However, existing LAFA methods are mostly based on Neural Networks (NN) that are difficult to interpret in practice. In this work, we introduce a tree distillation method for learning an interpretable policy from NN-based LAFA networks. We validated our method through both a simulation and an empirical EMA dataset on forecasting daily alcohol consumption. In both cases, we find that we can meaningfully reduce the number of items acquired at each occasion with minimal loss in accuracy. Networks (NN) that are difficult to interpret in practice. In this work, we introduce a tree distillation method for learning an interpretable policy from NN-based LAFA networks. We validated our method through both a simulation and an empirical EMA dataset on forecasting daily alcohol consumption. In both cases, we find that we can meaningfully reduce the number of items acquired at each occasion with minimal loss in accuracy.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Applications (stat.AP); Methodology (stat.ME); Machine Learning (stat.ML) |
| Cite as: | arXiv:2610.07452 [cs.LG] |
| (or arXiv:2610.07452v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.07452 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yunni Qu [view email]
[v1]
Mon, 5 Oct 2026 21:57:45 UTC (3,532 KB)
来源：arXiv：cs.LG · arxiv.org