scholarly journals Smart Predict-and-Optimize for Hard Combinatorial Optimization Problems

2020 ◽  
Vol 34 (02) ◽  
pp. 1603-1610 ◽  
Author(s):  
Jayanta Mandi ◽  
Emir Demirovi? ◽  
Peter J. Stuckey ◽  
Tias Guns
Keyword(s):  
Combinatorial Optimization ◽  
Objective Function ◽  
Large Scale ◽  
Optimization Problem ◽  
Optimization Problems ◽  
Machine Learning Techniques ◽  
Scheduling Problems ◽  
Combinatorial Optimization Problems ◽  
Main Challenge ◽  
Linear Programming Problems

Combinatorial optimization assumes that all parameters of the optimization problem, e.g. the weights in the objective function, are fixed. Often, these weights are mere estimates and increasingly machine learning techniques are used to for their estimation. Recently, Smart Predict and Optimize (SPO) has been proposed for problems with a linear objective function over the predictions, more specifically linear programming problems. It takes the regret of the predictions on the linear problem into account, by repeatedly solving it during learning. We investigate the use of SPO to solve more realistic discrete optimization problems. The main challenge is the repeated solving of the optimization problem. To this end, we investigate ways to relax the problem as well as warm-starting the learning and the solving. Our results show that even for discrete problems it often suffices to train by solving the relaxation in the SPO loss. Furthermore, this approach outperforms the state-of-the-art approach of Wilder, Dilkina, and Tambe. We experiment with weighted knapsack problems as well as complex scheduling problems, and show for the first time that a predict-and-optimize approach can successfully be used on large-scale combinatorial optimization problems.

scholarly journals A Methodology to Find the Elementary Landscape Decomposition of Combinatorial Optimization Problems

2011 ◽  
Vol 19 (4) ◽  
pp. 597-637 ◽  
Author(s):  
Francisco Chicano ◽  
L. Darrell Whitley ◽  
Enrique Alba
Keyword(s):  
Combinatorial Optimization ◽  
Objective Function ◽  
Optimization Problem ◽  
Optimization Problems ◽  
Quadratic Assignment Problem ◽  
Combinatorial Optimization Problem ◽  
Quadratic Assignment ◽  
Neighborhood Structure ◽  
Combinatorial Optimization Problems ◽  
Elementary Landscape

A small number of combinatorial optimization problems have search spaces that correspond to elementary landscapes, where the objective function f is an eigenfunction of the Laplacian that describes the neighborhood structure of the search space. Many problems are not elementary; however, the objective function of a combinatorial optimization problem can always be expressed as a superposition of multiple elementary landscapes if the underlying neighborhood used is symmetric. This paper presents theoretical results that provide the foundation for algebraic methods that can be used to decompose the objective function of an arbitrary combinatorial optimization problem into a sum of subfunctions, where each subfunction is an elementary landscape. Many steps of this process can be automated, and indeed a software tool could be developed that assists the researcher in finding a landscape decomposition. This methodology is then used to show that the subset sum problem is a superposition of two elementary landscapes, and to show that the quadratic assignment problem is a superposition of three elementary landscapes.

Space Contraction and Partition Algorithm for Combinatorial Optimization

2011 ◽  
Vol 421 ◽  
pp. 559-563
Author(s):  
Yong Chao Gao ◽  
Li Mei Liu ◽  
Heng Qian ◽  
Ding Wang
Keyword(s):  
Combinatorial Optimization ◽  
Optimization Problem ◽  
Optimization Problems ◽  
Solution Space ◽  
Search Space ◽  
Small Scale ◽  
Optimal Solutions ◽  
Solution Quality ◽  
Intelligent Algorithms ◽  
Combinatorial Optimization Problems

The scale and complexity of search space are important factors deciding the solving difficulty of an optimization problem. The information of solution space may lead searching to optimal solutions. Based on this, an algorithm for combinatorial optimization is proposed. This algorithm makes use of the good solutions found by intelligent algorithms, contracts the search space and partitions it into one or several optimal regions by backbones of combinatorial optimization solutions. And optimization of small-scale problems is carried out in optimal regions. Statistical analysis is not necessary before or through the solving process in this algorithm, and solution information is used to estimate the landscape of search space, which enhances the speed of solving and solution quality. The algorithm breaks a new path for solving combinatorial optimization problems, and the results of experiments also testify its efficiency.

scholarly journals Combinatorial optimization by simulating adiabatic bifurcations in nonlinear Hamiltonian systems

2019 ◽  
Vol 5 (4) ◽  
pp. eaav2372 ◽  
Author(s):  
Hayato Goto ◽  
Kosuke Tatsumura ◽  
Alexander R. Dixon
Keyword(s):  
Combinatorial Optimization ◽  
Hamiltonian Systems ◽  
Large Scale ◽  
Optimization Problems ◽  
Approximate Solutions ◽  
Combinatorial Optimization Problems ◽  
Bifurcation Phenomena ◽  
Field Programmable ◽  
Max Cut Problem ◽  
Adiabatic Optimization

Combinatorial optimization problems are ubiquitous but difficult to solve. Hardware devices for these problems have recently been developed by various approaches, including quantum computers. Inspired by recently proposed quantum adiabatic optimization using a nonlinear oscillator network, we propose a new optimization algorithm simulating adiabatic evolutions of classical nonlinear Hamiltonian systems exhibiting bifurcation phenomena, which we call simulated bifurcation (SB). SB is based on adiabatic and chaotic (ergodic) evolutions of nonlinear Hamiltonian systems. SB is also suitable for parallel computing because of its simultaneous updating. Implementing SB with a field-programmable gate array, we demonstrate that the SB machine can obtain good approximate solutions of an all-to-all connected 2000-node MAX-CUT problem in 0.5 ms, which is about 10 times faster than a state-of-the-art laser-based machine called a coherent Ising machine. SB will accelerate large-scale combinatorial optimization harnessing digital computer technologies and also offer a new application of computational and mathematical physics.

Matched Field Processing Based on the Simulated Annealing

2013 ◽  
Vol 651 ◽  
pp. 879-884
Author(s):  
Qi Wang ◽  
Ying Min Wang ◽  
Yan Ni Gou
Keyword(s):  
Global Optimization ◽  
Simulated Annealing ◽  
Combinatorial Optimization ◽  
Mediterranean Sea ◽  
Optimization Problem ◽  
Optimization Problems ◽  
Combinatorial Optimization Problem ◽  
Matched Field Processing ◽  
Combinatorial Optimization Problems ◽  
Passive Localization

The matched field processing (MFP) for localization usually needs to match all the replica fields in the observation sea with the received fields, and then find the maximum peaks in the matched results, so how to find the maximum in the results effectively and quickly is a problem. As known the classical simulated annealing (CSA) which has the global optimization capability is used widely for combinatorial optimization problems. For passive localization the position of the source can be recognized as a combinatorial optimization problem about range and depth, so a new matched field processing based on CSA is proposed. In order to evaluate the performance of this method, the normal mode was used to calculate the replica field. Finally the algorithm was evaluated by the dataset in the Mediterranean Sea in 1994. Comparing to the conventional matched field passive localization (CMFP), it can be conclude that the new one can localize optimum peak successfully where the output power of CMFP is maximum, meanwhile it is faster than CMFP.

Study on Free Search for Combinatorial Optimization Problem

2013 ◽  
Vol 411-414 ◽  
pp. 1904-1910
Author(s):  
Kai Zhong Jiang ◽  
Tian Bo Wang ◽  
Zhong Tuan Zheng ◽  
Yu Zhou
Keyword(s):  
Combinatorial Optimization ◽  
Optimization Problem ◽  
Feasible Solution ◽  
Optimization Problems ◽  
Search Algorithm ◽  
Search Process ◽  
Standard Data ◽  
Combinatorial Optimization Problems ◽  
Free Search

An algorithm based on free search is proposed for the combinatorial optimization problems. In this algorithm, a feasible solution is converted into a full permutation of all the elements and a transformation of one solution into another solution can be interpreted the transformation of one permutation into another permutation. Then, the algorithm is combined with intersection elimination. The discrete free search algorithm greatly improves the convergence rate of the search process and enhances the quality of the results. The experiment results on TSP standard data show that the performance of the proposed algorithm is increased by about 2.7% than that of the genetic algorithm.

scholarly journals Estimates of objective function minimum for solving linear fractional unconstrained combinatorial optimization problems on arrangements

2021 ◽  
pp. 32-36
Author(s):  
Tetiana Barbolina
Keyword(s):  
Combinatorial Optimization ◽  
Objective Function ◽  
Numerical Experiments ◽  
Polynomial Algorithm ◽  
Optimization Problems ◽  
Finite Sequence ◽  
Feasible Set ◽  
Combinatorial Optimization Problems ◽  
Unconstrained Problems

The paper is devoted to the study of one class of Euclidean combinatorial optimization problems — combinatorial optimization problems on the general set of arrangements with linear fractional objective function and without additional (non-combinatorial) constraints. The paper substantiates the improvement of the polynomial algorithm for solving the specified class of problems. This algorithm foresees solving a finite sequence of linear unconstrained problems of combinatorial optimization on arrangements. The modification of the algorithm is based on the use of estimates of the objective function on the feasible set. This allows to exclude some of the problems from consideration and reduce the number of problems to be solved. The numerical experiments confirm the practical efficiency of the proposed approach.

A Research on Flowshop Scheduling Problems With Column Generation

2012 ◽  
Author(s):  
H. Tanohata ◽  
T. Kaihara ◽  
N. Fujii
Keyword(s):  
Combinatorial Optimization ◽  
Local Search ◽  
Column Generation ◽  
Optimization Problems ◽  
New Method ◽  
Duality Gap ◽  
Scheduling Problems ◽  
Flowshop Scheduling ◽  
Feasible Schedule ◽  
Combinatorial Optimization Problems

Column generation is a method to calculate lowerbound for combinatorial optimization problems, although a feasible schedule is generally obtained with the upperbound. Therefore, in this paper, a new method is proposed to solve the flowshop scheduling problems with column generation, which is composed of the local search and duality gap termination condition. The neighborhood of the local search is composed of columns, and the method is applied in column generation to improve the upperbound and lowerbound. The effectiveness of the proposed method is verified by computational experiments.

Sign in / Sign up

Export Citation Format

Share Document