On the use of fitness landscape features in meta-learning based algorithm selection for the quadratic assignment problem

DOI10.1016/J.TCS.2019.10.033MaRDI QIDQ2285657zbMATH OpenOpenAlexWikidataFDO

Authors Augusto Dantas, Aurora Pozo

Publication date 8 January 2020

Published in Theoretical Computer Science (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/j.tcs.2019.10.033

zbMATH Keywords

combinatorial optimizationmeta-learningalgorithm selectionfitness landscape analysis

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Problem solving in the context of artificial intelligence (heuristics, search strategies, etc.) (68T20) Approximation methods and heuristics in mathematical programming (90C59) Combinatorial optimization (90C27) Discrete location and assignment (90B80)

Recommendations

• Summary of algorithm selection problems based on meta-learning
• Measuring instance difficulty for combinatorial optimization problems
• Fitness landscape analysis and metaheuristics efficiency
• scientific article; zbMATH DE number 1699999
• A novel feature-based approach to characterize algorithm performance for the traveling salesperson problem

Cites work

• scientific article; zbMATH DE number 714537 (Why is no real title available?)
• A new genetic algorithm for the quadratic assignment problem
• An improved annealing scheme for the QAP
• Assignment Problems and the Location of Economic Activities
• Breakout local search for the quadratic assignment problem
• Comparison of iterative searches for the quadratic assignment problem
• Continuous lunches are free plus the design of optimal optimization algorithms
• Iterated local search for the quadratic assignment problem
• MaxSAT by improved instance-specific algorithm configuration
• Metalearning
• QAPLIB - a quadratic assignment problem library
• Random forests
• Recent advances for the quadratic assignment problem with special emphasis on instances that are difficult for meta-heuristic methods
• SATzilla: portfolio-based algorithm selection for SAT
• SUNNY: a lazy portfolio approach for constraint solving
• Scikit-learn: machine learning in Python
• Solving large quadratic assignment problems on computational grids
• The Reactive Tabu Search
• The algorithm selection competitions 2015 and 2017
• The online performance estimation framework: heterogeneous ensemble learning for data streams
• The quadratic assignment problem. Theory and algorithms

Cited in

(6)

• Summary of algorithm selection problems based on meta-learning
• Migration-based algorithm library enrichment for constrained multi-objective optimization and applications in algorithm selection
• Machine learning at the service of meta-heuristics for solving combinatorial optimization problems: a state-of-the-art
• Discovering the suitability of optimisation algorithms by learning from evolved instances
• New variable-length data compression scheme for solution representation of meta-heuristics
• A landscape-based analysis of fixed temperature and simulated annealing

Describes a project that uses

Uses Software

• SUNNY
• sunny-cp
• Scikit
• SATzilla
• QAPLIB

This page was built for publication: On the use of fitness landscape features in meta-learning based algorithm selection for the quadratic assignment problem

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2285657)