Transcript Evolutionary Algorithms

Evolutionary Algorithms
Simon M. Lucas
The basic idea
• Initialise a random population of
individuals
repeat {
evaluate
select
vary (e.g. mutate or crossover)
} until satisfied, or timeup
Idea: fitness evolves over time
What to evolve?
• Parameter vectors for parametric solution
• Computational machines
– E.g. Finite state machines
– Neural networks
– Computer programs
• Expression trees
• OO programs
Current Research
• Evolving game playing agents
– Interesting links between species learning
– And lifetime learning
• Hybrid algorithms
– Evolution with temporal difference learning
Links with Biology
•
•
•
•
•
Evolving sequence classifiers
Cell models?
Gene regulatory networks?
Artificial immune systems?
Other ideas?