Swarm intelligence and evolutionary computation : theory, advances and applications in machine learning and deep learning

Swarm intelligence and evolutionary computation theory, advances and applications in machine learning and deep learning

This book aims at providing theoretical knowledge in the application of swarm intelligence and evolutionary computation including several recent meta-heuristic algorithms and also providing practical emerging applications in machine learning and deep learning.

Detalles Bibliográficos
Otros Autores: Kouziokas, Georgios N., author (author)
Formato: Libro electrónico
Idioma:Inglés
Publicado: Boca Raton, Florida : CRC Press [2023]
Materias:
Swarm intelligence.
Evolutionary computation.
Deep learning (Machine learning)
Ver en Biblioteca Universitat Ramon Llull:https://discovery.url.edu/permalink/34CSUC_URL/1im36ta/alma991009809021906719
Tabla de Contenidos:
  • Cover
  • Title Page
  • Copyright Page
  • Preface
  • Table of Contents
  • 1. Computational optimization
  • Computational optimization
  • Introduction
  • Optimization methods
  • Gauss-Newton method
  • Quasi-Newton method
  • Gradient-based optimization
  • Steepest or gradient descent algorithm
  • Conjugate gradient algorithms
  • Optimizers for machine learning
  • Stochastic gradient descent
  • Stochastic gradient descent with momentum
  • Levenberg-Marquardt algorithm
  • Scaled conjugate gradient algorithm
  • Adagrad
  • RMSProp
  • Adadelta
  • Adam optimizer
  • Non-gradient methods
  • References
  • 2. Evolutionary Computation and Genetic Algorithm
  • Evolutionary strategy
  • Genetic algorithm
  • Initialization
  • Selection methods
  • Tournament selection
  • Linear ranking selection
  • Proportionate roulette wheel selection
  • Exponential ranking selection
  • Crossover (recombination) operators
  • Crossover operators for binary encoding
  • Crossover operators for real-coded genetic algorithms
  • Mutation operators
  • Gaussian mutation
  • Cauchy mutation
  • Diversity mutation
  • Lévy flight mutation
  • Power mutation
  • Termination conditions
  • Fitness limit
  • Maximum number of generations
  • Maximum stall time
  • Maximum runtime
  • Best fitness value
  • Adaptive genetic algorithm
  • Differential evolution
  • Mutation
  • Crossover
  • Selection
  • Chaotic differential evolution
  • Differential evolution example
  • Ackley function approximation
  • References
  • 3. Swarm Intelligence and Particle Swarm Optimization
  • Particle swarm optimization algorithm
  • Hyper-parameters
  • Acceleration coefficients
  • Inertia weight
  • Stopping criteria
  • Maximum generation number
  • Maximum stall time
  • Maximum runtime
  • Best fitness value
  • Population convergence
  • Fitness convergence
  • Swarm topologies
  • Global or fully connected
  • Local or ring topology.
  • Von Neumann
  • Star topology
  • Mesh topology
  • Random topology
  • Tree or hierarchical topology
  • Dynamic or adaptive topologies
  • Boundary handling approaches
  • Hyperbolic method
  • Infinity or invisible wall
  • Nearest or boundary or absorb
  • Random
  • Random-half
  • Periodic
  • Exponential
  • Mutation
  • Reflect methods
  • Random damping
  • PSO with mutation
  • Gaussian mutation PSO
  • Cauchy mutation PSO
  • Michalewicz non-uniform mutation
  • Chaotic PSO with Michalewicz mutation
  • Random mutation PSO
  • Constant mutation PSO
  • Stagnant mutation
  • Quantum PSO
  • Delta well quantum PSO
  • Harmonic quantum PSO
  • Multi-objective PSO
  • Geometric PSO
  • PSO in neural network optimization
  • PSO as weight optimizer
  • PSO as topology optimizer
  • Swarm optimization examples
  • Sphere function
  • Griewank function
  • Rastrigin function
  • Ackley function
  • Schwefel function
  • References
  • 4. Ant Colony Optimization and Artificial Bee Colony
  • Ant colony optimization (ACO)
  • Introduction
  • Ant system (AS)
  • Ant colony system (ACS)
  • Rank-based ant system (RB-AS)
  • Max-min ant system (MMAS)
  • Population-based ACO
  • Artificial bee colony (ABC)
  • Bee colony foraging behavior
  • ABC algorithm
  • Selection methods
  • Boundary handling approaches
  • References
  • 5. Cuckoo Search and Bat Swarm Algorithm
  • Cuckoo search
  • Cuckoo breeding behavior and Lévy flights
  • Cuckoo search algorithm
  • Cuckoo search variants
  • Chaotic cuckoo search
  • Discrete binary cuckoo search
  • Hybrid self-adaptive cuckoo search
  • Bat algorithm
  • Bat algorithm inspiration
  • Bat movement
  • Bat algorithm variants
  • Binary bat algorithm
  • Chaotic bat algorithm
  • 1st Chaotic bat algorithm
  • 2nd Chaotic bat algorithm
  • 3rd Chaotic bat algorithm
  • Self-adaptive bat algorithm
  • Step-control mechanism
  • Mutation mechanism.
  • Bat algorithm with double mutation
  • Time factor modification
  • Cauchy mutation operator modification
  • Gaussian mutation operator
  • References
  • 6. Firefly Algorithm, Harmony Search and Cat Swarm Algorithm
  • Firefly algorithm
  • Firefly algorithm variants
  • Firefly algorithm with Lévy flights
  • Chaotic firefly algorithms
  • Harmony search algorithm
  • Harmony search example
  • Harmony search variants
  • Improved harmony search algorithm
  • Chaotic harmony search
  • Cat swarm Optimization
  • Cat swarm algorithm
  • Basic description of Cat Swarm Algorithm
  • Cat algorithm variants
  • Binary discrete Cat algorithm
  • Improved cat swarm optimization
  • References
  • 7. Grey Wolf, Whale and Grasshopper Optimization
  • Grey wolf optimization
  • Encircling prey
  • Hunting
  • Attacking prey (exploitation)
  • Search for prey (exploration)
  • Grey wolf algorithm variants
  • Binary grey wolf optimization
  • Grey wolf with Lévy flight
  • Whale optimization algorithm
  • Encircling prey
  • Bubble-net attacking strategy (exploitation phase)
  • Whale optimization variants
  • Whale optimization with Lévy flight
  • Binary whale optimization algorithm
  • Grasshopper optimization algorithm
  • Grasshopper optimization variants
  • Chaotic grasshopper optimization algorithm
  • Improved grasshopper optimization algorithm
  • References
  • 8. Machine Learning Optimization Applications
  • Artificial neural networks
  • Weight optimization of a neural network
  • Topology optimization of a neural network
  • Neural network training with PSO, ACO, GA
  • Experimental setup
  • Genetic algorithm parameters
  • PSO parameters
  • ACO parameters
  • Experimental results
  • Feature selection with swarm intelligence and genetic algorithm
  • Problem definition
  • Data analysis in machine learning
  • Energy consumption dataset
  • Data Pre-processing
  • Normalization.
  • Processing dataset outliers
  • Cost-based feature selection with swarm intelligence
  • Correlation-based feature selection with swarm intelligence
  • Experimental setup
  • Genetic algorithm
  • GA algorithm parameters
  • Genetic algorithm results
  • Geometric PSO
  • GPSO parameters
  • GPSO results
  • Chaotic harmony search
  • Algorithm parameters
  • Chaotic harmony search results
  • Chaotic Cuckoo Search
  • Algorithm parameters
  • Chaotic Cuckoo Search results
  • Evolutionary algorithm
  • EA parameters
  • EA results
  • Predictions with reduced features, SVM and random forest
  • Crime forecasting with PSO-SVM
  • References
  • 9. Swarm and Evolutionary Intelligence in Deep Learning
  • Deep LSTM and Bi-LSTM networks
  • Deep CNN (Convolutional Neural Networks)
  • CNN and LSTM optimization
  • Topology optimization
  • Weight optimization
  • Experiments
  • Bi-LSTM optimization
  • Dataset
  • Objective function
  • Experimental setup
  • Genetic algorithm parameters
  • Adaptive w-PSO parameters
  • Bidirectional LSTM training parameters
  • CNN optimization
  • CNN - PSO model
  • Evaluation metrics
  • Covid-19 chest X-ray dataset
  • Experimental results
  • CNN without PSO
  • CNN optimized with PSO
  • References
  • Index.

Ejemplares similares