Sparse Recovery with Oversampling Ratio Greater than One Half
arXiv:1706.080301.2
Originality Highly original
AI Analysis
It addresses a fundamental theoretical limit in sparse recovery, showing feasibility beyond the previously assumed bound.
The paper proves that sparse recovery is theoretically possible with oversampling ratio greater than one half and proposes two modified HTP algorithms achieving this.
Solving sparse recovery problem with high oversampling ratio is hard. We show that it is theoretically possible and we propose two modified HTP algorithms with such performances.