Genetic Programming: 25th European Conference, EuroGP 2022, Held as Part of EvoStar 2022, Madrid, Spain, April 20–22, 2022, Proceedings (Lecture Notes in Computer Science)

Preface
Organization
Contents
Long Presentations
Evolving Adaptive Neural Network Optimizers for Image Classification
	1 Introduction
	2 Background
	3 Adaptive AutoLR
		3.1 Grammar
		3.2 Fitness Function
	4 Experimental Study
		4.1 Evolutionary Runs
		4.2 Evolutionary Results
		4.3 Benchmark
		4.4 Fashion-MNIST
		4.5 CIFAR-10
	5 Conclusion
	References
Combining Geometric Semantic GP with Gradient-Descent Optimization
	1 Introduction
	2 Related Works
	3 Gradient Descent GSGP
		3.1 Geometric Semantic GP
		3.2 Adam Algorithm
		3.3 GSGP Hybridized with Gradient Descent
	4 Experimental Settings
		4.1 Dataset
		4.2 Experimental Study
	5 Experimental Results
	6 Conclusions
	References
One-Shot Learning of Ensembles of Temporal Logic Formulas for Anomaly Detection in Cyber-Physical Systems
	1 Introduction
	2 Related Work
	3 Background: Signal Temporal Logic
	4 Problem Statement
	5 Methodology
	6 Experimental Evaluation
		6.1 Datasets and Preprocessing
		6.2 Procedure and Evaluation Metrics
		6.3 Results
	7 Conclusions
	References
Multi-objective Genetic Programming with the Adaptive Weighted Splines Representation for Symbolic Regression
	1 Introduction
		1.1 Research Objectives
	2 Background
		2.1 Model Complexity and Generalisation
		2.2 Genetic Programming with Adaptive Weighted Splines
	3 Proposed Method
		3.1 Multi-objective Fitness Function
		3.2 Non-dominated Sorting Genetic Algorithm II
		3.3 Combining Multi-objective Optimization with the Adaptive Weighted Splines
	4 Experiment Settings
		4.1 Benchmark Methods
		4.2 Benchmark Problems
		4.3 Parameter Settings
	5 Results and Analysis
		5.1 Comparisons of Hypervolume Indicator
		5.2 Analyses of Fronts
		5.3 Visualizations and Analyses of GPSR Models in GP-AWS-PP
	6 Conclusions and Future Work
	References
SLUG: Feature Selection Using Genetic Algorithms and Genetic Programming
	1 Introduction
	2 SLUG
	3 Data
	4 Methods
	5 Experimental Setup
	6 Results
		6.1 Regular Classification Tasks
		6.2 Gametes Classification Tasks
	7 Discussion
	8 Conclusion and Future Work
	References
Evolutionary Design of Reduced Precision Levodopa-Induced Dyskinesia Classifiers
	1 Introduction
	2 LID-Classifier Design
		2.1 Clinical Study Data
		2.2 Data Preprocessing
		2.3 Classifier Model
		2.4 Classifier Training
	3 Experimental Setup
		3.1 Experiments
		3.2 CGP Setup
		3.3 CGPcoASFP Setup
		3.4 Time of Stabilization of LID-Classifier Evolution
	4 Results
		4.1 Experiment 1: Comparisons of CGP and CGPcoASFP
		4.2 Experiment 2: Comparisons of Data Representations
		4.3 Experiment 3: Hardware Characteristics of Evolved Classifiers
	5 Conclusions
	References
Using Denoising Autoencoder Genetic Programming to Control Exploration and Exploitation in Search
	1 Introduction
	2 Related Work
	3 Denoising Autoencoder LSTMs
		3.1 Autoencoder LSTMs
		3.2 Suggesting a New Denoising Strategy: Levenshtein Edit
		3.3 Training Procedure
		3.4 Sampling with Syntax Control
	4 Experiments
		4.1 Experimental Setup
		4.2 Performance Results
		4.3 The Influence of Denoising on Search
	5 Conclusions
	References
Program Synthesis with Genetic Programming: The Influence of Batch Sizes
	1 Introduction
	2 Lexicase Selection in GP-Based Program Synthesis
	3 Methodology
		3.1 Benchmark Problems
		3.2 Grammars
		3.3 Selection Method
	4 Experiments and Results
		4.1 Influence on Selection Pressure
		4.2 Analysis of Success Rates and Generalization
	5 Conclusions
	References
Genetic Programming-Based Inverse Kinematics for Robotic Manipulators
	1 Introduction
	2 Related Work
	3 Genetic Programming-Based Inverse Kinematics
		3.1 Fitness Functions to Model the IK Problem
		3.2 Cooperative Coevolutionary GP for Inverse Kinematics
	4 Experimental Evaluation
		4.1 Data Processing
		4.2 Experiment Setup
		4.3 Preliminary Experiments
		4.4 Advanced Experiments
		4.5 Discussion
	5 Conclusion and Future Work
	References
On the Schedule for Morphological Development of Evolved Modularpg Soft Robots
	1 Introduction and Related Works
	2 Background: Voxel-Based Soft Robots
		2.1 VSR Morphology
		2.2 VSR Controller
	3 Development of VSRs
		3.1 Representations for the Development Function
		3.2 Evolution of the Development Function
	4 Experimental Evaluation
		4.1 Results and Discussion
	5 Concluding Remarks
	References
An Investigation of Multitask Linear Genetic Programming for Dynamic Job Shop Scheduling
	1 Introduction
	2 Background
		2.1 Dynamic Job Shop Scheduling
		2.2 Linear Genetic Programming
		2.3 Related Work
	3 Multitask LGPHH
		3.1 Multi-factorial LGPHH
		3.2 Multitask Multi-population LGPHH
	4 Experiment Design
		4.1 Multitask DJSS Scenarios
		4.2 Comparison Methods
	5 Results and Discussion
		5.1 Test Performance
		5.2 Example Program Analysis
	6 Conclusion
	References
Cooperative Co-evolution and Adaptive Team Composition for a Multi-rover Resource Allocation Problem
	1 Introduction
	2 Cooperative Coevolutionary and Team Composition
	3 Cooperative Co-evolutionary Algorithms with Limited Team Composition Update
	4 The Multi-rover Resource Selection Problem
	5 Results
		5.1 Experimental Setting
		5.2 Fixed vs. Adaptive Methods for Team Composition Update
		5.3 Dynamics of Adapting the Number of Team Agents to Update
		5.4 Sensitivity of Meta-parameters
	6 Conclusion
	References
Short Presentations
Synthesizing Programs from Program Pieces Using Genetic Programming and Refinement Type Checking
	1 Introduction
	2 Method
		2.1 Program Synthesis Model
		2.2 Genetic Programming Algorithm
		2.3 Refinement Types and LiquidHaskell
		2.4 Refinement Types Fitness Function
	3 Experiments and Results
		3.1 Program Synthesis Problems
		3.2 Experimental Setup
		3.3 Results
		3.4 Threats to Validity
	4 Related Work
	5 Conclusions
	References
Creating Diverse Ensembles for Classification with Genetic Programming and Neuro-MAP-Elites
	1 Introduction
	2 Background
		2.1 Linear Genetic Programming
		2.2 Map-Elites
		2.3 Variational Autoencoders (VAEs)
		2.4 Ensemble Classifiers
	3 Our LGP Implementation
	4 Neuro MAP-Elites
		4.1 Mine Solutions
		4.2 VAE Training
		4.3 MAP Elites with Encoder
	5 Experiment Setup
		5.1 Dataset Selection
		5.2 Standard Machine Learning Classifiers
		5.3 Map-Elites Classifiers
	6 Results
		6.1 VAE Efficacy
		6.2 Diversity Comparison
		6.3 Ensemble Accuracy
	7 Discussion
	References
Evolving Monotone Conjunctions in Regimes Beyond Proved Convergence
	1 Introduction
		1.1 Monotone Conjunctions and Representation
		1.2 Related Work and Motivation
	2 Computational Models Relevant to Our Work
		2.1 Evolutionary Algorithms and Evolving Programs
		2.2 Supervised Machine Learning and Evolvability
	3 The Learning Problem that We Study
		3.1 A Related Algorithm: The Swapping Algorithm
	4 Implementation
		4.1 Setting the Parameters q and
		4.2 Guessing a Good Value for the Tolerance t
		4.3 Successful Executions
	5 Experimental Results and Discussion
		5.1 Details on the Experimental Setup
		5.2 High-Level Summary of Results
		5.3 Details on the Convergence When p=0.4
		5.4 Further Details on the Experiments of Every (p, 69640972 c86418188 ) Pair Tested
		5.5 Discussion
	6 Conclusions
	References
Accurate and Interpretable Representations of Environments with Anticipatory Learning Classifier Systems
	1 Introduction
	2 Related Works
		2.1 Principles of ALCS
		2.2 ALCS and Non-determinism
	3 Behavioral Enhanced Anticipatory Classifier System
		3.1 Enhancing PEP into EPE
		3.2 Coupling EPE with Behavioral Sequences
		3.3 Enhancing the Behavioral Sequences
	4 Performance in Maze Environments
		4.1 Experimental Protocol
		4.2 Metrics
		4.3 Performance
	5 Discussion
	6 Conclusion
	References
Exploiting Knowledge from Code to Guide Program Search
	1 Introduction
	2 Related Work
	3 GitHub Code Corpus
		3.1 Software Metrics
		3.2 Descriptive Analysis of the Code Corpus
	4 Experiments
		4.1 Experimental Setup
		4.2 Results and Discussion
	5 Conclusion
	References
Multi-objective Genetic Programming for Explainable Reinforcement Learning
	1 Introduction
	2 Related Work
	3 Explainable Reinforcement Learning Using GP
	4 The Experiments
		4.1 GP Representations
		4.2 Benchmarks and Evaluation
		4.3 Baselines
	5 Experimental Results
		5.1 Quantitative Analysis
		5.2 Dealing with the Local Minimum Trap
	6 Conclusion and Further Work
	References
Permutation-Invariant Representation of Neural Networks with Neuron Embeddings
	1 Introduction
	2 Related Work
	3 Method
	4 Experiments
		4.1 Training from Random Initialization
		4.2 Compression Ability
		4.3 Cross-model Compatibility
	5 Conclusion
	References
Author Index