Artificial Immune System (AIS) |
good for upscaled and complex optimization problems
parallelizable
good for finding global optima while avoiding becoming stuck in local optima
able to learn from search space and self-improve accordingly
robust to noisy environments
ensures diversity of solutions
good for dynamic problems and pattern recognition
|
can be computationally complex and expensive
lack of solid theoretical foundation
challenging understanding and analysis
complex and time-consuming parameter tuning
applicable to a relatively narrow range of problems
|