Evolutionary Computation in Combinatorial Optimization: 23rd European Conference, EvoCOP 2023 Held as Part of EvoStar 2023 Brno, Czech Republic, April 12–14, 2023 Proceedings

Preface
Organization
Contents
Fairer Comparisons for Travelling Salesman Problem Solutions Using Hash Functions
	1 Introduction
	2 Collision Analysis of the Fitness Function on the TSP
		2.1 Too Many Collisions for the Fitness Function
		2.2 Distribution of Collisions over Fitness Values
	3 Hash Functions for a Reliable Comparison
		3.1 Existing Hash Functions
		3.2 The Proposed Function
		3.3 Comparative Study
	4 Revisiting Some Metaheuristics with Hash Functions
		4.1 Cycling Analysis
		4.2 Convergence Speed
		4.3 Applying Hash Function on Metaheuristics
	5 Discussion and Conclusion
	References
Application of Adapt-CMSA to the Two-Echelon Electric Vehicle Routing Problem with Simultaneous Pickup and Deliveries
	1 Introduction
	2 Problem Description
	3 Adapt-CMSA for the 2E-EVRP-SPD
	4 Experimental Evaluation
	5 Conclusion and Outlook
	References
Real-World Vehicle Routing Using Adaptive Large Neighborhood Search
	1 Introduction
	2 Mathematical Model
	3 Our Approach
		3.1 Two-Stage Minimization
		3.2 Insertion Optimization
		3.3 Removal Optimization
	4 Experimental Evaluation
		4.1 Li and Lim Benchmark Instances
		4.2 Real-World Instances
	5 Conclusion
	References
A Multilevel Optimization Approach for Large Scale Battery Exchange Station Location Planning
	1 Introduction
	2 Related Work
	3 The Multi-period Battery Swapping Station Location Problem
	4 Multilevel Refinement Algorithm
	5 Computational Results
	6 Conclusion and Future Work
	References
A Memetic Algorithm for Deinterleaving Pulse Trains
	1 Introduction
	2 Deinterleaving Markov Processes: Formal Background
		2.1 Interleaved Markov Generative Process
		2.2 Penalized Maximum Likelihood Score
	3 Problem Settings and Motivation for this Work
		3.1 Decomposable Score for Estimating Processes Optimal Order
		3.2 A Combinatorial Problem in the Space of Partitions
	4 A Memetic Algorithm for Alphabet Partitioning
		4.1 General Framework
		4.2 Initialisation
		4.3 Tabu Search Procedure
		4.4 Greedy Likelihood-Based Crossover Operator
	5 Experiments and Computational Results
		5.1 Experimental Condition and Reference Algorithm
		5.2 Experiments on Synthetic Datasets
		5.3 Experiments on Electronic Warfare Datasets
	6 Conclusions
	References
Application of Negative Learning Ant Colony Optimization to the Far from Most String Problem
	1 Introduction
		1.1 Organization of the Paper
	2 Far from Most String Problem
		2.1 Integer Linear Programming Model
		2.2 Computational Complexity and Previous Work
	3 The Proposed Algorithm
		3.1 Construction of a Solution
		3.2 Solving Sub-instances
		3.3 Update of the Pheromone Values
	4 Different Objective Functions
	5 Experimental Evaluation
		5.1 Benchmark Sets
		5.2 Algorithm Tuning
		5.3 Results
	6 Conclusions and Outlook
	References
Monte Carlo Tree Search with Adaptive Simulation: A Case Study on Weighted Vertex Coloring
	1 Introduction
	2 Related Works on the WVCP
		2.1 Local Search Algorithms
		2.2 Constructive Heuristics
	3 MCTS with Adaptive Simulation Strategy
		3.1 Main Scheme
		3.2 Adaptive Simulation Strategy Framework
	4 Operator Selectors
		4.1 Neural Network Selector
		4.2 Classic Fitness-Based Selectors
	5 Experimentation
		5.1 Experimental Settings and Benchmark Instances
		5.2 Adaptive Operator Selection During the Search
		5.3 Performance Comparisons on the Different Benchmark Instances
	6 Conclusions
	References
Evolutionary Strategies for the Design of Binary Linear Codes
	1 Introduction
	2 Background
		2.1 Binary Linear Codes
		2.2 Boolean Functions
	3 Related Works
	4 Evolutionary Strategy Algorithm
		4.1 Solutions Encoding and Search Space
		4.2 Fitness Function
		4.3 Rank-Preserving Mutation and Crossover
	5 Experiments
		5.1 Experimental Setting
		5.2 Results
		5.3 Solutions Diversity
	6 Conclusions and Future Work
	References
A Policy-Based Learning Beam Search for Combinatorial Optimization
	1 Introduction
	2 Related Work
	3 Policy-Based Learning Beam Search
		3.1 Loss Functions
		3.2 Bootstrapping
	4 Neural Network Architecture
	5 Case Study: Longest Common Subsequence Problem
	6 Experimental Evaluation
	7 Conclusions and Future Work
	References
Cooperative Coevolutionary Genetic Programming Hyper-Heuristic for Budget Constrained Dynamic Multi-workflow Scheduling in Cloud Computing
	1 Introduction
	2 Background and Related Work
		2.1 Background
		2.2 Related Work
	3 Problem Model
	4 Methodology
		4.1 Budget Constrained Cooperative Coevolution Genetic Programming Hyper-Heuristic
	5 Experiments
		5.1 Experiment Design
		5.2 Experiment Results
		5.3 Analysis
	6 Conclusions
	References
OneMax Is Not the Easiest Function for Fitness Improvements
	1 Introduction
		1.1 Our Result
	2 Preliminaries and Definitions
		2.1 The Algorithm: SA-(1, )-EA
		2.2 The Benchmarks: OneMax and Dynamic BinVal
		2.3 Tools
	3 Main Proof
		3.1 Sketch of Proof
		3.2 Proof Details
	4 Simulations
	5 Conclusion
	References
The Cost of Randomness in Evolutionary Algorithms: Crossover can Save Random Bits
	1 Introduction
		1.1 Our Contributions
		1.2 Related Work
	2 Preliminaries and Analysis Tools
	3 Cost of Randomness in Mutation-Based EAs
	4 Cost of Randomness with Crossover
	5 Detailed Analysis for the (2+1) GA
	6 Conclusions and Future Work
	References
Multi-objectivization Relaxes Multi-funnel Structures in Single-objective NK-landscapes
	1 Introduction
	2 Single-objective Landscape
		2.1 Single-objective Optimization
		2.2 Basin of Attraction
		2.3 Local Optima Networks
		2.4 Funnel
	3 Multi-objective Landscape
		3.1 Multi-objective Optimization
		3.2 Pareto Local Optimal Solutions Networks
	4 NK- And MNK-landscape Problems
		4.1 NK-landscape Problem
		4.2 MNK-landscape Problem
	5 Proposed Method: Multi-Objectivized Local Search
		5.1 Motivation
		5.2 Algorithm
	6 Experimental Setup
		6.1 Landscape Analysis
		6.2 Algorithm Benchmarks
	7 Experimental Results and Discussions
		7.1 Landscape Analysis
		7.2 Algorithm Benchmarks
	8 Conclusions
	References
Decision/Objective Space Trajectory Networks for Multi-objective Combinatorial Optimisation
	1 Introduction
	2 Multi-objective Combinatorial Optimisation
		2.1 Definitions
		2.2 Multi-objective Evolutionary Algorithms
	3 Search Trajectory Networks
		3.1 Definitions
		3.2 Network Metrics
		3.3 Network Visualisation
	4 Experimental Setup
		4.1 Benchmark Problems
		4.2 Parameter Setting
		4.3 Reproducibility
	5 Results
		5.1 Small Instances
		5.2 Large Instances
	6 Conclusions
	References
On the Effect of Solution Representation and Neighborhood Definition in AutoML Fitness Landscapes
	1 Introduction
	2 Related Work
	3 AutoML Fitness Landscape
		3.1 Solution Representation and Neighborhood
		3.2 Fitness Function
	4 Characterization of the Fitness Landscapes
	5 Exploratory Fitness Landscape Analysis
	6 Results and Discussion
		6.1 Fitness Distance Correlation
		6.2 Dispersion Metric
		6.3 Neutrality Rate
	7 Conclusions and Future Work
	References
Author Index