ReLU networks as surrogate models in mixed-integer linear programs

被引:90
|
作者
Grimstad, Bjarne [1 ]
Andersson, Henrik [2 ]
机构
[1] Solut Seeker AS, Gaustadalleen 21, N-0349 Oslo, Norway
[2] Norwegian Univ Sci & Technol, Dept Ind Econ & Technol Management, NO-7491 Trondheim, Norway
关键词
Deep neural networks; ReLU networks; Mixed-Integer linear programming; Surrogate modeling; Regression; SEQUENTIAL DESIGN STRATEGY; GLOBAL OPTIMIZATION; NEURAL-NETWORKS;
D O I
10.1016/j.compchemeng.2019.106580
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
We consider the embedding of piecewise-linear deep neural networks (ReLU networks) as surrogate models in mixed-integer linear programming (MILP) problems. A MILP formulation of ReLU networks has recently been applied by many authors to probe for various model properties subject to input bounds. The formulation is obtained by programming each ReLU operator with a binary variable and applying the big-M method. The efficiency of the formulation hinges on the tightness of the bounds defined by the big-M values. When ReLU networks are embedded in a larger optimization problem, the presence of output bounds can be exploited in bound tightening. To this end, we devise and study several bound tightening procedures that consider both input and output bounds. Our numerical results show that bound tightening may reduce solution times considerably, and that small-sized ReLU networks are suitable as surrogate models in mixed-integer linear programs. (C) 2019 Elsevier Ltd. All rights reserved.
引用
收藏
页数:15
相关论文
共 50 条
  • [21] A mixed-integer linear programming approach to the reduction of genome-scale metabolic networks
    Roehl, Annika
    Bockmayr, Alexander
    BMC BIOINFORMATICS, 2017, 18
  • [22] A mixed-integer linear programming approach to the reduction of genome-scale metabolic networks
    Annika Röhl
    Alexander Bockmayr
    BMC Bioinformatics, 18
  • [23] Mixed-integer linear programming models for batch sterilization of packaged-foods plants
    R. Simpson
    A. Abakarov
    Journal of Scheduling, 2013, 16 : 59 - 68
  • [24] Learning Presolver Selection for Mixed-Integer Linear Programming
    Song, Wentao
    Gu, Naijie
    2024 16TH INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND COMPUTING, ICMLC 2024, 2024, : 635 - 641
  • [25] Testing copositivity via mixed-integer linear programming
    Anstreicher, Kurt M.
    LINEAR ALGEBRA AND ITS APPLICATIONS, 2021, 609 : 218 - 230
  • [26] Efficient configuration of heterogeneous multistatic sonar networks: A mixed-integer linear programming approach
    Thuillier, Owein
    Le Josse, Nicolas
    Olteanu, Alexandru-Liviu
    Sevaux, Marc
    Tanguy, Herve
    COMPUTERS & OPERATIONS RESEARCH, 2024, 167
  • [27] Mixed integer linear models for the optimization of dynamical transport networks
    Geissler, Bjoern
    Kolb, Oliver
    Lang, Jens
    Leugering, Guenter
    Martin, Alexander
    Morsi, Antonio
    MATHEMATICAL METHODS OF OPERATIONS RESEARCH, 2011, 73 (03) : 339 - 362
  • [28] Mixed-integer linear programming models for batch sterilization of packaged-foods plants
    Simpson, R.
    Abakarov, A.
    JOURNAL OF SCHEDULING, 2013, 16 (01) : 59 - 68
  • [29] Mixed integer linear models for the optimization of dynamical transport networks
    Björn Geißler
    Oliver Kolb
    Jens Lang
    Günter Leugering
    Alexander Martin
    Antonio Morsi
    Mathematical Methods of Operations Research, 2011, 73 : 339 - 362
  • [30] Global optimization of mixed-integer nonlinear programs with SCIP 8
    Bestuzheva, Ksenia
    Chmiela, Antonia
    Mueller, Benjamin
    Serrano, Felipe
    Vigerske, Stefan
    Wegscheider, Fabian
    JOURNAL OF GLOBAL OPTIMIZATION, 2025, 91 (02) : 287 - 310