No Cross-Validation Required: An Analytical Framework for Regularized Mixed-Integer Problems

被引:1
|
作者
Soleimani, Behrad [1 ]
Khamidehi, Behzad [2 ]
Sabbaghian, Maryam [3 ]
机构
[1] Univ Maryland, Dept ECE, College Pk, MD 20742 USA
[2] Univ Toronto, Dept ECE, Toronto, ON M5S 1A1, Canada
[3] Univ Tehran, Dept ECE, Tehran 1417466191, Iran
关键词
Mixed-integer programming; regularization; alternating method; penalty function; RAT selection; OPTIMIZATION; CONVERGENCE; ASSIGNMENT; SELECTION;
D O I
10.1109/LCOMM.2020.3013377
中图分类号
TN [电子技术、通信技术];
学科分类号
0809 ;
摘要
This letter develops a method to obtain the optimal value for the regularization coefficient in a general mixed-integer problem (MIP). This approach eliminates the cross-validation performed in the existing penalty techniques to obtain a proper value for the regularization coefficient. We obtain this goal by proposing an alternating method to solve MIPs. First, via regularization, we convert the MIP into a more mathematically tractable form. Then, we develop an iterative algorithm to update the solution along with the regularization (penalty) coefficient. We show that our update procedure guarantees the convergence of the algorithm. Moreover, assuming the objective function is continuously differentiable, we derive the convergence rate, a lower bound on the value of regularization coefficient, and an upper bound on the number of iterations required for the convergence. We use a radio access technology (RAT) selection problem in a heterogeneous network to benchmark the performance of our method. Simulation results demonstrate near-optimality of the solution and consistency of the convergence behavior with obtained theoretical bounds.
引用
收藏
页码:2868 / 2872
页数:5
相关论文
共 50 条
  • [31] Using mixed-integer programming to solve power grid blackout problems
    Bienstock, Daniel
    Mattia, Sara
    DISCRETE OPTIMIZATION, 2007, 4 (01) : 115 - 141
  • [32] Modelling practical lot-sizing problems as mixed-integer programs
    Belvaux, G
    Wolsey, LA
    MANAGEMENT SCIENCE, 2001, 47 (07) : 993 - 1007
  • [33] Solving elementary shortest-path problems as mixed-integer programs
    Michael Drexl
    Stefan Irnich
    OR Spectrum, 2014, 36 : 281 - 296
  • [34] Time-Domain Decomposition for Mixed-Integer Optimal Control Problems
    Falk M. Hante
    Richard Krug
    Martin Schmidt
    Applied Mathematics & Optimization, 2023, 87
  • [35] Transfer Learning for Mixed-Integer Resource Allocation Problems in Wireless Networks
    Shen, Yifei
    Shi, Yuanming
    Zhang, Jun
    Letaief, Khaled B.
    ICC 2019 - 2019 IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS (ICC), 2019,
  • [36] The Value of Randomized Solutions in Mixed-Integer Distributionally Robust Optimization Problems
    Delage, Erick
    Saif, Ahmed
    INFORMS JOURNAL ON COMPUTING, 2022, 34 (01) : 333 - 353
  • [37] An adaptive patch approximation algorithm for bicriteria convex mixed-integer problems
    Diessel, Erik
    OPTIMIZATION, 2022, 71 (15) : 4321 - 4366
  • [38] An exact penalty global optimization approach for mixed-integer programming problems
    S. Lucidi
    F. Rinaldi
    Optimization Letters, 2013, 7 : 297 - 307
  • [39] Concurrent processing of mixed-integer non-linear programming problems
    Ostermark, Ralf
    KYBERNETES, 2009, 38 (06) : 970 - 993
  • [40] Solution of Chance-Constrained Mixed-Integer Nonlinear Programming Problems
    Esche, Erik
    Mueller, David
    Werk, Sebastian
    Grossmann, Ignacio E.
    Wozny, Guenter
    26TH EUROPEAN SYMPOSIUM ON COMPUTER AIDED PROCESS ENGINEERING (ESCAPE), PT A, 2016, 38A : 91 - 96