A Supervised Learning Approach Involving Active Subspaces for an Efficient Genetic Algorithm in High-Dimensional Optimization Problems.
Nicola DemoMarco TezzeleGianluigi RozzaPublished in: SIAM J. Sci. Comput. (2021)
Keyphrases
- high dimensional
- optimization problems
- genetic algorithm
- supervised learning
- evolutionary algorithm
- metaheuristic
- low dimensional
- high dimensional data
- dimensionality reduction
- training samples
- unsupervised learning
- fitness function
- parameter space
- feature space
- cost function
- multi objective
- dimension reduction
- data points
- ant colony optimization
- nearest neighbor
- traveling salesman problem
- optimization methods
- crossover operator
- high dimensionality
- similarity search
- high dimensional feature spaces
- fuzzy logic
- active learning
- reinforcement learning
- training data
- neural network
- supervised machine learning
- learning algorithm
- nsga ii
- unlabeled data
- objective function
- semi supervised
- training set
- variable selection
- generalization error
- multiple instance learning
- multi dimensional
- lower dimensional
- high dimensional spaces
- labeled data
- semi supervised learning