A globally convergent algorithm for nonconvex optimization based on block coordinate update

Yangyang Xu University of Alabama Wotao Yin UCLA

Numerical Analysis and Scientific Computing Optimization and Control mathscidoc:1703.25020

Journal of Scientific Computing, 2017
Nonconvex optimization arises in many areas of computational science and engineering. However, most nonconvex optimization algorithms are only known to have local convergence or subsequence convergence properties. In this paper, we propose an algorithm for nonconvex optimization and establish its global convergence (of the whole sequence) to a critical point. In addition, we give its asymptotic convergence rate and numerically demonstrate its efficiency. In our algorithm, the variables of the underlying problem are either treated as one block or multiple disjoint blocks. It is assumed that each non-differentiable component of the objective function, or each constraint, applies only to one block of variables. The differentiable components of the objective function, however, can involve multiple blocks of variables together. Our algorithm updates one block of variables at a time by minimizing a certain prox-linear surrogate, along with an extrapolation to accelerate its convergence. The order of update can be either deterministically cyclic or randomly shuffled for each cycle. In fact, our convergence analysis only needs that each block be updated at least once in every fixed number of iterations. We show its global convergence (of the whole sequence) to a critical point under fairly loose conditions including, in particular, the Kurdyka–Łojasiewicz condition, which is satisfied by a broad class of nonconvex/nonsmooth applications. These results, of course, remain valid when the underlying problem is convex.We apply our convergence results to the coordinate descent iteration for non-convex regularized linear regression, as well as amodified rank-one residue iteration for nonnegative matrix factorization. We show that both applications have global convergence. Numerically,we tested our algorithm on nonnegative matrix and tensor factorization problems, where random shuffling clearly improves the chance to avoid low-quality local solutions.
Nonconvex optimization, Nonsmooth optimization, Block coordinate descent, Kurdyka-Lojasiewicz inequality
[ Download ] [ 2017-03-14 09:27:45 uploaded by yangyangxu ] [ 199 downloads ] [ 0 comments ]
  • Published online first
@inproceedings{yangyang2017a,
  title={A globally convergent algorithm for nonconvex optimization based on block coordinate update},
  author={Yangyang Xu, and Wotao Yin},
  url={http://archive.ymsc.tsinghua.edu.cn/pacm_paperurl/20170314092745408269689},
  booktitle={Journal of Scientific Computing},
  year={2017},
}
Yangyang Xu, and Wotao Yin. A globally convergent algorithm for nonconvex optimization based on block coordinate update. 2017. In Journal of Scientific Computing. http://archive.ymsc.tsinghua.edu.cn/pacm_paperurl/20170314092745408269689.
Please log in for comment!
 
 
Contact us: office-iccm@tsinghua.edu.cn | Copyright Reserved