Thuật ngữ AI
Từ điển đầy đủ về Trí tuệ nhân tạo
Multi-Objective PSO
Extension of the Particle Swarm Optimization algorithm designed to solve optimization problems involving multiple, often conflicting objective functions, seeking an optimal set of Pareto solutions.
Global Guide
Selection mechanism in multi-objective PSO to choose one or more particles from the external archive to guide the movement of other particles toward the Pareto front.
Personal Best (pbest)
Historically best position of a specific particle, evaluated based on multi-objective criteria such as non-dominance or belonging to the current Pareto front.
Solution Diversity
Measure of the spatial dispersion of solutions on the Pareto front, crucial in multi-objective PSO to avoid premature convergence to a limited region of the search space.
Multi-Objective Velocity Vector
Position update component of a particle in multi-objective PSO, often adapted to manage trade-offs between different objectives when moving in the search space.
Hypervolume (HV)
Performance indicator measuring the volume of objective space covered by a set of solutions, used to evaluate the quality and convergence of a multi-objective PSO algorithm toward the true Pareto front.
MOPSO (Multi-Objective Particle Swarm Optimization)
Acronym referring to particle swarm optimization algorithms specifically adapted to handle multi-objective optimization problems.
Sigma Method
Global guide selection technique in MOPSO that uses sigma values to direct particles toward promising areas of the Pareto front based on the angle of objective vectors.
Multi-Criteria Decision
Final process of selecting a single solution from the set of Pareto solutions provided by multi-objective PSO, based on the decision maker's preferences.
Multi-Objective Combinatorial Optimization
Application of multi-objective PSO to problems where the search space is discrete, requiring specific adaptations for the representation and update of particle positions.