Low-Rank Tensor Recovery
2022
During data acquisition and transmission, some entries of data are missing, which will degrade the performance of subsequent data processing. Missing component analysis, also named matrix completion, can recover the missing data based on the low-rank assumption. However, with the emergence of high-order data, traditional methods directly tackle the high-order data by rearranging it into a matrix, which inevitably lose some structural information. As a generation of matrix completion, tensor completion is proposed to recover the missing entries of high-order data.
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
58
References
2
Citations
NaN
KQI