MTC: Multiresolution Tensor Completion from Partial and Coarse Observations

MTC: Multiresolution Tensor Completion from Partial and Coarse Observations
复制标题

MTC:部分和粗略观测的多分辨率张量补全

DOI:
10.1145/3447548.3467261
复制
发表时间:
2021
期刊:
KDD '21: Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining
影响因子:
--
通讯作者:
Sun, Jimeng
Sun, Jimeng
中科院分区:
--
文献类型:
--
作者:
Yang, Chaoqi;Singh, Navjot;Xiao, Cao;Qian, Cheng;Solomonik, Edgar;Sun, Jimeng

文献摘要

参考文献

被引文献

相似文献

现有的张量补全公式大多依赖于单个张量的部分观测。然而,从真实世界数据中提取的张量通常更加复杂,这是因为:(I)部分观察:只有一小部分张量元素可用。(2)粗略观察:一些张量模式只呈现粗略和聚合的模式(例如,每月摘要而不是每日报告)。本文给出了张量的一个子集和一些聚集/粗略的观测值(沿着一个或多个模式),并试图用低阶因式分解恢复原始的细粒度张量。我们建立了一个耦合张量补全问题,并提出了一个有效的多分辨率张量补全模型(MTC)来解决该问题。我们的MTC模型探索张量模式的性质,并利用分辨率的层次递归地初始化优化设置,并使用交替最小二乘对耦合系统进行优化。MTC可确保较低的计算和空间复杂度。我们在两个新冠肺炎相关的时空张量上对我们的模型进行了评估。实验表明,MTC在仅有5%的细粒度观测值的情况下,可以提供65.20%和75.79%的张量完成适应度(PoF),比最佳基线提高了27.96%。为了评估学习的低阶因子,我们还设计了一个张量预测任务,用于每日和累积的疾病病例预测,其中MTC在POF中实现了50%的POF,在最佳基线上实现了30%的相对改善。
Existing tensor completion formulation mostly relies on partial observations from a single tensor. However, tensors extracted from real-world data often are more complex due to: (i) Partial observation: Only a small subset of tensor elements are available. (ii) Coarse observation: Some tensor modes only present coarse and aggregated patterns (e.g., monthly summary instead of daily reports). In this paper, we are given a subset of the tensor and some aggregated/coarse observations (along one or more modes) and seek to recover the original fine-granular tensor with low-rank factorization. We formulate a coupled tensor completion problem and propose an efficient Multi-resolution Tensor Completion model (MTC) to solve the problem. Our MTC model explores tensor mode properties and leverages the hierarchy of resolutions to recursively initialize an optimization setup, and optimizes on the coupled system using alternating least squares. MTC ensures low computational and space complexity. We evaluate our model on two COVID-19 related spatio-temporal tensors. The experiments show that MTC could provide 65.20% and 75.79% percentage of fitness (PoF) in tensor completion with only 5% fine granular observations, which is 27.96% relative improvement over the best baseline. To evaluate the learned low-rank factors, we also design a tensor prediction task for daily and cumulative disease case predictions, where MTC achieves 50% in PoF and 30% relative improvements over the best baseline.
DOI: 10.1007/978-3-030-47436-2_65
发表时间: 2020-04-17
期刊: Advances in Knowledge Discovery and Data Mining
影响因子: --
作者:
Almutairi FM;Kanatsoulis CI;Sidiropoulos ND
通讯作者: Sidiropoulos ND
DOI: --
发表时间: 2020-02
期刊: ArXiv
影响因子: --
作者:
Jung Yeon Park;K. T. Carr;Stephan Zhang;Yisong Yue;Rose Yu
通讯作者: Jung Yeon Park;K. T. Carr;Stephan Zhang;Yisong Yue;Rose Yu
具有模式层次结构的多分辨率张量分解
DOI: 10.1145/2532169
发表时间: 2014
期刊: ACM Trans. Knowl. Discov. Data
影响因子: --
作者:
C. Schifanella;K. Candan;M. Sapino
通讯作者: M. Sapino
低阶典型张量分解的自适应代数多重网格算法
DOI: 10.1137/110855934
发表时间: 2011
期刊: SIAM J. Sci. Comput.
影响因子: --
作者:
H. Sterck;Killian Miller
通讯作者: Killian Miller
Prema:从多个聚合视图恢复有原则的张量数据
DOI: 10.1109/jstsp.2021.3056918
发表时间: 2021
影响因子: 7.5
作者:
Almutairi, Faisal M.;Kanatsoulis, Charilaos I.;Sidiropoulos, Nicholas D.
通讯作者: Sidiropoulos, Nicholas D.