Hybridizing Cartesian Genetic Programming and Harmony Search for adaptive feature construction in supervised learning problems.
Andoni ElolaJavier Del SerMiren Nekane BilbaoCristina PerfectoEnrique AlexandreSancho Salcedo-SanzPublished in: Appl. Soft Comput. (2017)
Keyphrases
- feature construction
- learning problems
- genetic programming
- harmony search
- supervised learning
- genetic algorithm
- learning algorithm
- metaheuristic
- differential evolution
- machine learning
- learning tasks
- evolutionary algorithm
- simulated annealing algorithm
- optimization algorithm
- machine learning algorithms
- semi supervised learning
- evolutionary computation
- bp neural network
- fitness function
- semi supervised
- kernel methods
- unsupervised learning
- pac learning
- optimization method
- sample complexity
- hill climbing
- decision trees
- neural network
- simulated annealing
- optimization problems
- training examples
- active learning
- feature selection
- training data
- artificial neural networks
- classification rules
- reinforcement learning
- fuzzy logic
- search space
- class labels
- multiple instance learning
- unlabeled data
- feature space
- relational learning
- expert systems
- training set