L1 minimization using ADMM #6
Description
This work provides implementations of algorithms in the paper: ALTERNATING DIRECTION ALGORITHMS FOR ℓ1-PROBLEMS IN COMPRESSIVE SENSING by Junfeng Yang and Yin Zhang.
Algorithms:
- Solve basis pursuit (1.3)
- Solve basis pursuit with inequality constraints (1.4)
- Solve basis pursuit denoising (1.5)
- Solve l_1/l_1 problem (1.6)
We also need to implement corresponding non-negative counterparts.
Evaluation:
- Recovery results from data corrupted by impulsive noise only
- Recovery results from data corrupted by white noise only
- Recovery results from data corrupted by both white and impulsive noise
- Relative error vs optimality for noiseless and noisy data