{
  "id": "1407.0753",
  "version": "v3",
  "published": "2014-07-03T00:29:25.000Z",
  "updated": "2014-12-03T00:59:38.000Z",
  "title": "Global convergence of splitting methods for nonconvex composite optimization",
  "authors": [
    "Guoyin Li",
    "Ting Kei Pong"
  ],
  "comment": "Simple sufficient conditions guaranteeing boundedness of sequence have been added",
  "categories": [
    "math.OC",
    "cs.LG",
    "math.NA",
    "stat.ML"
  ],
  "abstract": "We consider the problem of minimizing the sum of a smooth function $h$ with a bounded Hessian, and a nonsmooth function. We assume that the latter function is a composition of a proper closed function $P$ and a surjective linear map $\\cal M$, with the proximal mappings of $\\tau P$, $\\tau > 0$, simple to compute. This problem is nonconvex in general and encompasses many important applications in engineering and machine learning. In this paper, we examined two types of splitting methods for solving this nonconvex optimization problem: alternating direction method of multipliers and proximal gradient algorithm. For the direct adaptation of the alternating direction method of multipliers, we show that, if the penalty parameter is chosen sufficiently large and the sequence generated has a cluster point, then it gives a stationary point of the nonconvex problem. We also establish convergence of the whole sequence under an additional assumption that the functions $h$ and $P$ are semi-algebraic. Furthermore, we give simple sufficient conditions to guarantee boundedness of the sequence generated. These conditions can be satisfied for a wide range of applications including the least squares problem with the $\\ell_{1/2}$ regularization. Finally, when $\\cal M$ is the identity so that the proximal gradient algorithm can be efficiently applied, we show that any cluster point is stationary under a slightly more flexible constant step-size rule than what is known in the literature for a nonconvex $h$.",
  "revisions": [
    {
      "version": "v2",
      "updated": "2014-08-02T13:09:55.000Z",
      "title": "Splitting methods for nonconvex composite optimization",
      "abstract": "We consider the problem of minimizing the sum of a smooth function $h$ with a bounded Hessian, and a nonsmooth function. We assume that the latter function is a composition of a proper closed function $P$ and a surjective linear map $\\mathcal{M}$, with the proximal mappings of $\\tau P$, $\\tau > 0$, simple to compute. This problem is nonconvex in general and encompasses many important applications in engineering and machine learning. In this paper, we examined two types of splitting methods for solving this nonconvex optimization problem: alternating direction method of multipliers and proximal gradient algorithm. For the direct adaptation of the alternating direction method of multipliers, we show that, if the penalty parameter is chosen sufficiently large and the sequence generated has a cluster point, then it gives a stationary point of the nonconvex problem. We also establish convergence of the whole sequence under an additional assumption that the functions $h$ and $P$ are semi-algebraic. Furthermore, when $\\mathcal{M}$ is the identity so that the proximal gradient algorithm can be efficiently applied, we show that any cluster point is stationary under a slightly more flexible constant step-size rule than what is known in the literature for a nonconvex $h$.",
      "comment": null,
      "journal": null,
      "doi": null
    },
    {
      "version": "v3",
      "updated": "2014-12-03T00:59:38.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "nonconvex composite optimization",
      "splitting methods",
      "proximal gradient algorithm",
      "alternating direction method",
      "cluster point"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable",
      "adsabs": "2014arXiv1407.0753L"
    }
  }
}