2016 Fiscal Year Annual Research Report
Bayesian Tensor Models for Multiway Structural Data: A Theoretical Study and Applications
| Project/Area Number |
15K15955
|
|---|
| Research Institution | Institute of Physical and Chemical Research |
|---|
Principal Investigator |
ZHAO QIBIN 国立研究開発法人理化学研究所, 脳科学総合研究センター, 研究員 (30599618)
|
|---|
| Project Period (FY) |
2015-04-01 – 2017-03-31
|
| Keywords | Tensor Factorization / Bayesian Inference / Tensor Completion / Tensor Denoising |
|---|
| Outline of Annual Research Achievements |
We proposed Bayesian method for Tucker decomposition and Tucker completion. By employing group sparsity priors over factor matrices and cores, our method can automatically learn the multilinear rank from given observed tensor data.
We developed tensor denoising algorithm based on Bayesian low-rank tensor factorization, which can be used for image, video and MRI denoising. Furthermore, we apply Bayesian tensor methods to EEG signal analysis and human action recognition from video data.
|
Research Products
(3 results)
