18
X-Machines for Agent-Based Modeling: FLAME Perspectives
M o d e l
h y p o t h e s i s
M e a s u r e d
r e a l w o r l d
G e n e r a l
s o l u t i o n
E s t i m a t e d
r e a l w o r l d
D e d u c t i v e
m a n i p u l a t i o n
I n d u c t i v e
g e n e r a l i s a t i o n
I n d e p e n d e n t
v e r f i c a t i o n
D e d u c t i v e
s p e c i a l i s a t i o n
FIGURE 2.1: Scientific method. cf. [61].
2.1 Intelligent Agents
Evolutionary computation is a sub-field under artificial intelligence (AI)
research area, involving optimization to automatically solve difficult problems.
These contain the following:
• Always include an iterative process where models progressively update
their performance.
• Allow growth of given agent populations such that are internally modified based on performance.
• Processes can involve parallel processing.
• Mostly all processes are inspired by principles of natural evolution.
Evolutionary computation contains four sub-topics: genetic algorithms,
evolutionary programming, genetic programming and evolutionary strategies.
These are shown in detail in Figure 2.2.
Swarm optimization algorithms do not belong to this group, even if used as
one of the four approaches. Swarm optimization techniques are inspired from
insect colonies and involve large number of individuals working individually
to collectively solve the problem. For example, in Figure 1.6, ants could find
shortest possible routes to food sources by simply working together and leaving
pheromone trails for other ants [37]. These searches are constantly updated
depending on food availability and quality.
The focus of evolutionary computation research is mainly the algorithms
studying real systems, focusing on optimization and search problems. These
problems are difficult to solve and have high complexities, where evolutionary
algorithms can keep efficiency high at lower cost.
X-Machines for Agent-Based Modeling: FLAME Perspectives
M o d e l
h y p o t h e s i s
M e a s u r e d
r e a l w o r l d
G e n e r a l
s o l u t i o n
E s t i m a t e d
r e a l w o r l d
D e d u c t i v e
m a n i p u l a t i o n
I n d u c t i v e
g e n e r a l i s a t i o n
I n d e p e n d e n t
v e r f i c a t i o n
D e d u c t i v e
s p e c i a l i s a t i o n
FIGURE 2.1: Scientific method. cf. [61].
2.1 Intelligent Agents
Evolutionary computation is a sub-field under artificial intelligence (AI)
research area, involving optimization to automatically solve difficult problems.
These contain the following:
• Always include an iterative process where models progressively update
their performance.
• Allow growth of given agent populations such that are internally modified based on performance.
• Processes can involve parallel processing.
• Mostly all processes are inspired by principles of natural evolution.
Evolutionary computation contains four sub-topics: genetic algorithms,
evolutionary programming, genetic programming and evolutionary strategies.
These are shown in detail in Figure 2.2.
Swarm optimization algorithms do not belong to this group, even if used as
one of the four approaches. Swarm optimization techniques are inspired from
insect colonies and involve large number of individuals working individually
to collectively solve the problem. For example, in Figure 1.6, ants could find
shortest possible routes to food sources by simply working together and leaving
pheromone trails for other ants [37]. These searches are constantly updated
depending on food availability and quality.
The focus of evolutionary computation research is mainly the algorithms
studying real systems, focusing on optimization and search problems. These
problems are difficult to solve and have high complexities, where evolutionary
algorithms can keep efficiency high at lower cost.