We study the L1 minimization problem with additional box constraints. We motivate the problem with two different views of optimality considerations. We look into imposing such constraints in projected gradient techniques and propose a worst case linear time algorithm to perform such projections. We demonstrate the merits and effectiveness of our algorithms on synthetic as well as real experiments.
@article{arxiv.1010.0141,
title = {L1 Projections with Box Constraints},
author = {Mithun Das Gupta and Sanjeev Kumar and Jing Xiao},
journal= {arXiv preprint arXiv:1010.0141},
year = {2010}
}