eduzhai > Applied Sciences > Engineering >

GRATE Granular Recovery of Aggregated Tensor Data by Example

  • KanKan
  • (0) Download
  • 20210430
  • Save

... pages left unread,continue reading

Document pages: 20 pages

Abstract: In this paper, we address the challenge of recovering an accurate breakdownof aggregated tensor data using disaggregation examples. This problem ismotivated by several applications. For example, given the breakdown of energyconsumption at some homes, how can we disaggregate the total energy consumedduring the same period at other homes? In order to address this challenge, wepropose GRATE, a principled method that turns the ill-posed task at hand into aconstrained tensor factorization problem. Then, this optimization problem istackled using an alternating least-squares algorithm. GRATE has the ability tohandle exact aggregated data as well as inexact aggregation where someunobserved quantities contribute to the aggregated data. Special emphasis isgiven to the energy disaggregation problem where the goal is to provide energybreakdown for consumers from their monthly aggregated consumption. Experimentson two real datasets show the efficacy of GRATE in recovering more accuratedisaggregation than state-of-the-art energy disaggregation methods.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...