Statistical Mechanical Approach for the Theory of Compressed Sensing
| Project/Area Number |
24700007
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Fundamental theory of informatics
|
|---|
| Research Institution | Ibaraki University (2013-2015)
Tokyo Institute of Technology (2012) |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2012-04-01 – 2016-03-31
|
| Project Status |
Completed (Fiscal Year 2015)
|
| Budget Amount *help |
¥4,420,000 (Direct Cost: ¥3,400,000、Indirect Cost: ¥1,020,000)
Fiscal Year 2015: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2014: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2013: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2012: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
|
|---|
| Keywords | 圧縮センシング / 統計力学 / 統計物理学 / 高速疎データ再構成 / パラメータ最適化 |
|---|
| Outline of Final Research Achievements |
In the problem of compressed sensing, a significant algorithm called Approximate Message Passing (AMP) is widely-known, whose computational cost is small and convergence condition has been revealed theoretically. In deriving AMP, an assumption is made on the measurement matrix, which represents the process of signal measurement. In this project, we have attempted to construct generalized algorithm of AMP without such assumption, and have shown that such generalization can be achieved with the aid of formulation for the performance analysis of telecommunication system. We also investigated several topics on data sparsity: For example, we developed a sparse signal recovery algorithm of compressed sensing with less computational cost for spare measurement matrix. We also studied the sparsity-based learning algorithm for deep-learning.
|
Report
(5 results)
Research Products
(28 results)
