AI-ordlista
Den kompletta ordlistan över AI
Evolutionary Hybridization
Process of strategic integration of local search operators within an evolutionary algorithm to improve individuals after their creation by crossover and mutation.
Memetic Local Search
Optimization component selectively applied to individuals of an evolutionary population to locally refine their solutions and efficiently exploit the immediate neighborhood.
Lamarckian Strategy
Memetic approach where the individual improved by local search directly replaces the original in the gene pool, thus transmitting the acquired traits to the next generation.
Baldwinian Strategy
Memetic method where the improvement by local search does not modify the individual but adjusts its fitness, allowing the evolution of more adaptive genetic structures without direct alteration.
Algorithmic Meme
Unit of information or improvement schema transmitted between individuals by memetic operators, representing partial solutions or effective search heuristics.
Memetic Intensification
Optimization phase dedicated to the in-depth exploitation of promising regions of the solution space through repeated application of local search on the best individuals.
Memetic Diversification
Mechanism preserving genetic variability by limiting the application of local search or by introducing local minima escape operators to avoid premature convergence.
Memetic Self-Adaptation
Ability of a memetic algorithm to dynamically adjust its hybridization parameters, local search frequency, and operator strategies according to the problem characteristics.
Accelerated Memetic Convergence
Phenomenon of rapid improvement in solution quality resulting from the synergy between broad genetic exploration and targeted local exploitation of the best structures.
Memetic Hyper-heuristic
Meta-optimizer that dynamically selects and adapts different local search strategies and hybrid configurations according to the evolution phases and the characteristics of the fitness landscape.
Memetic Knowledge Transfer
Process of extracting and applying knowledge acquired during local searches to guide the future evolution of the population and improve the efficiency of genetic operators.
Memetic Selection
Mechanism for selecting individuals benefiting from local search based on multi-objective criteria combining fitness, genetic diversity, and estimated improvement potential.
Adaptive Neighborhood
Dynamic local search structure whose size and configuration evolve according to previous successes and the local topology of the solution space.
Memetic Complexity
Measure of the computational and temporal resources required by the memetic algorithm, influenced by the frequency and depth of local search applications.
Premature Local Convergence
Specific risk in memetic algorithms where excessive intensification of local search can lead to local minima before the complete exploration of the solution space.