arXiv · Statistics Machine Learning· Yunni Qu (Department of Computer Science, University of North Carolina at Chapel Hill), Bing Cai Kok (Department of Psychology and Neuroscience, University of North Carolina at Chapel Hill, School of Social Sciences, Nanyang Technological University, Singapore), Whitney Ringwald (Department of Psychology, University of Minnesota Twin Cities), Grant King (Department of Psychology, University of Michigan), Aidan Wright (Department of Psychology, University of Michigan), Kathleen Gates (Department of Psychology and Neuroscience, University of North Carolina at Chapel Hill), Junier Oliva (Department of Computer Science, University of North Carolina at Chapel Hill)·· 4 hr agoAI score22
用于成本高效时序预测的主动特征采集：树蒸馏提升可解释性并降低参与者负担
Active Feature Acquisition for Cost-Efficient Temporal Prediction with Reduced Participant Burden
AI brief
该研究提出一种树蒸馏方法，从基于神经网络的纵向主动特征采集（LAFA）网络中学习可解释的采集策略，在每个时间点动态选择要采集的条目子集以预测特定结果。方法在仿真和一个用于预测每日饮酒量的经验 EMA 数据集上验证，两种情况下都能显著减少每次采集的条目数，且精度损失极小。论文 arXiv:2610.07452，归属 cs.LG、cs.AI、stat.AP、stat.ME 与 stat.ML。
Source: arXiv · Statistics Machine Learning · arxiv.org