Learning with Submodular Functions: A Convex Optimization Perspective

被引:219
作者
Bach, Francis [1 ]
机构
[1] INRIA, Ecole Normale Super, Paris, France
来源
FOUNDATIONS AND TRENDS IN MACHINE LEARNING | 2013年 / 6卷 / 2-3期
基金
欧洲研究理事会;
关键词
D O I
10.1561/2200000039
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Submodular functions are relevant to machine learning for at least two reasons: (1) some problems may be expressed directly as the optimization of submodular functions and (2) the Lovasz extension of submodular functions provides a useful set of regularization functions for supervised and unsupervised learning. In this monograph, we present the theory of submodular functions from a convex analysis perspective, presenting tight links between certain polyhedra, combinatorial optimization and convex optimization problems. In particular, we show how submodular function minimization is equivalent to solving a wide variety of convex optimization problems. This allows the derivation of new efficient algorithms for approximate and exact submodular function minimization with theoretical guarantees and good practical performance. By listing many examples of submodular functions, we review various applications to machine learning, such as clustering, experimental design, sensor placement, graphical model structure learning or subset selection, as well as a family of structured sparsity- inducing norms that can be derived and used from submodular functions.
引用
收藏
页码:145 / 373
页数:233
相关论文
共 202 条
  • [1] Maximizing a class of submodular utility functions
    Ahmed, Shabbir
    Atamtuerk, Alper
    [J]. MATHEMATICAL PROGRAMMING, 2011, 128 (1-2) : 149 - 169
  • [2] Ahuja R. K., 1993, NETWORK FLOWS
  • [3] CONCAVITY OF CERTAIN MAPS ON POSITIVE DEFINITE MATRICES AND APPLICATIONS TO HADAMARD PRODUCTS
    ANDO, T
    [J]. LINEAR ALGEBRA AND ITS APPLICATIONS, 1979, 26 (AUG) : 203 - 241
  • [4] Babenko M., 2007, P INT C EXP ALG WAE
  • [5] Bach F., 2008, ARXIV08091493
  • [6] Bach F., 2011, ADV NEURAL INFORM PR
  • [7] Bach F., 2013, 00861118 HAL
  • [8] Bach F., 2013, 00757696V3 HAL
  • [9] Bach F., 2010, ADV NEURAL INFORM PR
  • [10] Optimization with Sparsity-Inducing Penalties
    Bach, Francis
    Jenatton, Rodolphe
    Mairal, Julien
    Obozinski, Guillaume
    [J]. FOUNDATIONS AND TRENDS IN MACHINE LEARNING, 2012, 4 (01): : 1 - 106