Genetic Programming: 23rd European Conference, EuroGP 2020, Held as Part of EvoStar 2020, Seville, Spain, April 15–17, 2020, Proceedings (Theoretical Computer Science and General Issues Book 12101)

This book constitutes the refereed proceedings of the 23rd European Conference on Genetic Programming, EuroGP 2020, held as part of Evo*2020, in Seville, Spain, in April 2020, co-located with the Evo*2020 events EvoCOP, EvoMUSART and EvoApplications. The 12 full papers and 6 short papers presented in this book were carefully reviewed and selected from 36 submissions. The papers cover a wide spectrum of topics, including designing GP algorithms for ensemble learning, comparing GP with popular machine learning algorithms, customising GP algorithms for more explainable AI applications to real-world problems. Preface Organization Contents Hessian Complexity Measure for Genetic Programming-Based Imputation Predictor Selection in Symbolic Regression with Incomplete Data 1 Introduction 2 Background 2.1 Missing Value Imputation 2.2 Model Complexity in GP 2.3 GP for Feature Selection 2.4 Symbolic Regression with Incomplete Data 3 The Proposed Method 3.1 The Overall System 3.2 Standard GP-Based Predictor Selection 3.3 GP-Based Predictor Selection with Feature Selection Pressure 3.4 The Proposed Method: GP-Based Predictor Selection with Model Complexity Pressure 4 Experiment Setup 5 Results and Discussions 5.1 Imputation Performance 5.2 Symbolic Regression Performance 5.3 The Number of Selected Predictors 6 Conclusions and Future Work References Seeding Grammars in Grammatical Evolution to Improve Search Based Software Testing 1 Introduction 2 Background and Related Work 2.1 Evolutionary Testing 2.2 SBST Techniques Benefitting from Seeding 3 Ariadne: GE-Based Test Data Generation 3.1 Grammatical Evolution 3.2 Grammar 4 Improved Grammar 4.1 Philosophy Behind the Proposed Changes 5 Experimental Results and Discussion 5.1 Experimental Setup 5.2 Detailed Analysis of Experiments 6 Conclusion and Future Work References Incremental Evolution and Development of Deep Artificial Neural Networks 1 Introduction 2 Related Work 3 Fast-DENSER 4 Incremental Development of Deep Neural Networks 5 Experimentation 5.1 Datasets 5.2 Experimental Setup 5.3 Experimental Results: Incremental Development 5.4 Experimental Results: Topology Analysis 5.5 Experimental Results: Generalisation of the Models 5.6 Discussion 6 Conclusions and Future Work References Investigating the Use of Geometric Semantic Operators in Vectorial Genetic Programming 1 Introduction 2 Panel Datasets in GP: Literature Review 3 Problem Description and the Datasets 3.1 Mosquito Abundance (P_Mosq) 3.2 Ventilation Flow (P_Physio) 4 Methodology 4.1 Vectorial Genetic Programming 4.2 Geometric Semantic Operators 5 Experiments 5.1 Experimental Settings 5.2 Experimental Results 6 Conclusions References Comparing Genetic Programming Approaches for Non-functional Genetic Improvement 1 Introduction 2 Genetic Improvement (GI) 2.1 Software Representations 2.2 Fitness Assessment 3 Genetic Programming (GP) 4 Experimental Setup 4.1 MiniSAT 4.2 Experimental Protocol 4.3 Search Processes 4.4 Filtering 5 Results and Discussion 5.1 Overall Training Results 5.2 Comparison of Approaches 5.3 Comparative Analysis 5.4 Research Questions 6 Conclusions References Automatically Evolving Lookup Tables for Function Approximation 1 Introduction 2 Background 2.1 Covariance Matrix Adaption - Evolution Strategy (CMA-ES) 2.2 Evolving Better Software Parameters 2.3 Investigating Evolving Better Software Parameters 3 Methods 3.1 CMA-ES Settings 3.2 Test Setup and Measurements 3.3 Fitness Function Design 4 Results 4.1 Run-Time Performance 4.2 Limitations 5 Conclusions and Outlook References Optimising Optimisers with Push GP 1 Introduction 2 Related Work 3 Methods 3.1 Push and Push GP 3.2 Evolving Population-Based Optimisers 3.3 Evaluation 4 Results 5 Conclusions References An Evolutionary View on Reversible Shift-Invariant Transformations 1 Introduction 2 Background 2.1 Shift-Invariant Transformations and Cellular Automata 2.2 Reversible CA 2.3 Marker CA 3 Optimizing Landscapes 3.1 Genotype Representation for Marker CA 3.2 Fitness Functions 4 Related Work 5 Experiments 5.1 Research Questions and Experimental Setting 5.2 Single-Objective Approach 5.3 Multi-objective Approach 5.4 Lexicographic Optimization 6 Conclusions and Future Work References Benchmarking Manifold Learning Methods on a Large Collection of Datasets 1 Introduction 2 Methods 2.1 Manifold Learning Methods 2.2 ManiGP - A New Manifold Learning Method Based on Genetic Programming 2.3 Datasets 2.4 Methodology of Comparison 3 Results 4 Conclusions References Ensemble Genetic Programming 1 Introduction 2 Related Work 3 Ensemble GP 3.1 M3GP 3.2 eGP Population Structure 3.3 eGP Fitness Functions 3.4 eGP Genetic Operators 4 Experimental Setup 4.1 Methods 4.2 Parameters 4.3 Datasets 5 Results 6 Discussion 7 Conclusions and Future Work References SGP-DT: Semantic Genetic Programming Based on Dynamic Targets 1 Introduction 2 Methodology 3 Related Work 4 Evaluation 4.1 Methods 4.2 Evaluation Setup 4.3 Results and Discussion 5 Conclusion References Effect of Parent Selection Methods on Modularity 1 Introduction 2 Related Work 3 Parent Selection Algorithms 3.1 Lexicase Selection 3.2 Tournament Selection 3.3 Fitness-Proportionate Selection 4 Push and the Evolution of Modularity 5 Reuse Metric 6 Experimental Set-Up 7 Results 8 Discussion 9 Conclusions and Future Work References Time Control or Size Control? Reducing Complexity and Improving Accuracy of Genetic Programming Models 1 Introduction 2 Background 2.1 Complexity in Genetic Programming 2.2 Evaluating Time Is More Than Measuring Size 2.3 Stabilising Evaluation Time Measurements 3 Experiments 3.1 Bloat Control Techniques 3.2 Test Problems 3.3 Configuration and Parameters 3.4 Initialising the Population 4 Results 4.1 Discussion 5 Conclusions and Future Work References Challenges of Program Synthesis with Grammatical Evolution 1 Introduction 2 Related Work 3 Methodology 3.1 Software Metrics 3.2 Program Synthesis Problems 3.3 GE Grammar and Fitness Function 4 Experiments and Discussion 4.1 Robustness of Reference Implementations: Part I 4.2 Robustness of Reference Implementations: Part II 4.3 Search Behavior of GE 4.4 Search for the Needle in a Haystack 5 Conclusions References Detection of Frailty Using Genetic Programming 1 Introduction 2 Methods 2.1 Data Source 2.2 Data Transformation 2.3 Learning from Imbalanced Data 3 Experiments 3.1 GP Parameter Setup 4 Results 4.1 GP Prediction Performance 4.2 Performance of Other Non-GP Classifiers 4.3 Feature Selection Comparison of GP and Chi-Square 5 Discussions and Conclusions References Is k Nearest Neighbours Regression Better Than GP? 1 Introduction 2 Geometric Semantic Genetic Programming 3 Random Vector Based Mutation 4 Experimental Study 4.1 Test Problems and Experimental Settings 4.2 Experimental Results: RVMGP vs GSGP 4.3 Experimental Results: RVMGP vs KNN Regression vs RF Regression 5 Conclusions and Future Work References Guided Subtree Selection for Genetic Operators in Genetic Programming for Dynamic Flexible Job Shop Scheduling 1 Introduction 2 Background 2.1 Dynamic Flexible Job Shop Scheduling 2.2 Genetic Programming Hyper-heuristic for DFJSS 3 The Proposed GP with Subtree Selection 3.1 The Occurrences of Features 3.2 The Importance of Subtrees 3.3 Subtree Selection 3.4 Summary 4 Experiment Design 4.1 Simulation Model 4.2 Parameter Settings 4.3 Comparison Design 5 Results and Discussions 5.1 Performance of Evolved Rules 5.2 The Probability Difference 5.3 The Occurrences of Features 5.4 Training Time 6 Conclusions and Future Work References Classification of Autism Genes Using Network Science and Linear Genetic Programming 1 Introduction 2 Methods 2.1 Data Collection 2.2 Human Molecular Interaction Network 2.3 Linear Genetic Programming Algorithm 2.4 Implementation Settings 3 Results 3.1 Properties of the HMIN 3.2 Best Classification Models 3.3 Assessment of Feature Importance 3.4 Evaluation of Autism-Gene Prioritization 3.5 Independent Validation of Autism-Gene Prediction 4 Discussion References Author Index