Download Motivating examples

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Polyploid wikipedia , lookup

DNA barcoding wikipedia , lookup

Philopatry wikipedia , lookup

Species distribution wikipedia , lookup

Hybrid (biology) wikipedia , lookup

Microevolution wikipedia , lookup

Koinophilia wikipedia , lookup

Transcript
24-03-2014
Multimodal optimization: speciation
E. Eiben and J. E. Smith, Introduction to Evolutionary
Computing, Springer, 2003 – chapter 5
Campbell, et al. 2008. Biology, 8th ed. Pearson
Benjamin Cummings.
Motivating examples
2
1
24-03-2014
Classification
„
Organisms
are
classified
into
a
hierarchical grouping of
closely
relates
organisms and that
progressively includes
more
and
more
organisms.
Species
„
The species is the basic biological unit
around which classifications are based.
„
However, what constitutes a species can be
difficult to define and there are multiple
definitions of species in use today.
2
24-03-2014
Early definition attempts (John Ray, 1627-1705)
„
A group of organisms that can interbreed and
produce viable, fertile offspring
Criticisms of the Biological Species Concept
1.
It applies only to sexually reproducing species.
2.
Distinguishing between species on the basis of
reproductive separation is problematic because
it can be difficult to determine how much
reproductive separation is needed to
distinguish between species.
3.
The definition refers only to current populations
and ignores the species status of ancestral
populations.
3
24-03-2014
Criticisms of the Biological Species Concept
„
Can a Great Dane
interbreed with a
Chihuahua?
Criticisms
„
Some species look and behave
differently
„
Can interbreed and have viable,
fertile offspring
„
Example: coyotes, wolves, dogs
4
24-03-2014
Criticisms
„
Normally, 25%
genetic diversity
implies different
species
„
However, human and
chimp genomes differ
by only 1.2%
Evolutionary Species Concept (G. Simpson, 1940)
„
“A single lineage of ancestor-descendant
populations that maintains its identity from
other such lineages and that has its own
evolutionary tendencies and historical fate.”
5
24-03-2014
Phylogenetic species concept
„
“an irreducible (basal) grouping of organisms
diagnosably distinct from other such
groupings and within which there is a
parental pattern of ancestry and descent.”
Species, adopted definition
„
A group of interbreeding individuals that
share a gene pool and are reproductively
isolated from other species.
6
24-03-2014
Speciation
Species: basic classification unit
Continuous lineage - information
passed through genes
Speciation: the mechanism by which
a new species is formed.
Two types of speciation are:
1.
2.
Allopatric speciation
Sympatric speciation
Allopatric speciation (extrinsic mechanism)
„
„
Allopatric (allo = other, patric = place)
Occurs when two populations are geographically
isolated from one another. They are separated
by distance or an impassable barrier.
7
24-03-2014
Speciation (extrinsic and intrinsic) mechanism
Sympatric speciation (intrinsic mechanism)
„
„
Sympatric (sym = same, patric = place)
New species develops when members of a
population develop a genetic difference that
prevents them from reproducing with members
of the original species.
8
24-03-2014
Speciation, summing up
„
As
new
species
evolve,
population
become
reproductively isolated i.e., member of 2 populations
cannot interbreed and produce fertile offspring
„
This tend to lead to phenotypic homogeneity within
species, and differences between species
„
Speciation can occur at widely different rates
Speciation in EAs
„
Niching does not prevent individuals in one
niche (maximum) to mate individuals in another
niche (maximum).
„
Speciation inspired EAs allow individuals to
mate only with similar ones
„
Similarity can be based either on
‰
‰
distance between representations, or on
matching between extra bits added to individual
representation
18
9
24-03-2014
Niching and Speciation
Fernando Lobo
University of Algarve
Speciation methods
• Deb and Goldberg proposed two mating restricting
schemes based on phenotypic and genotypic distances
between mating individuals.
• Idea: if the distance is closer than a parameter
, they
participate in the crossover operation. Otherwise another
individual is chosen at random as a candidate mate.
– This process continues until a proper mate is found. If
no such member exists, then a random individual from
the population is chosen as mate.
10
24-03-2014
Speciation methods
• Niching with
operator.
is implemented in the selection
• Mating restriction with
crossover operator.
is implemented in the
• Deb and Goldberg used
• Results on F1 and F2 were improved.
F1: equal peaks
F1(x) =
11
24-03-2014
F1 results
F1 with and without mating restriction
12
24-03-2014
Other speciation method
• Using tag bits (Spears, 1994)
• Tag bits are appended to every individual.
• Each species correspond to a particular configuration of
those bits.
• Mating is restricted to individuals containing the same set
of tag bits.
• Spears allowed the tag bits to be mutated.
Restricted Tournament Selection (RTS)
• RTS is a crowding-like method proposed by Harik (1995).
• RTS is a modification of tournament selection that restricts
an individual from competing with others that are very
different from it.
13
24-03-2014
Recalling Crowding
„
Attempts to distribute individuals evenly amongst niches
„
Deterministic Crowding:
‰
‰
‰
‰
‰
‰
randomly select parents p1, p2, and produce offspring o1, and o2
Evaluate the fitness value of offspring, f (o1) and f (o2), and
calculate their distances to parents, i.e., d(p1, o1), d(p1,o2), d(p2,
o1), and d(p2, c2).
If [d (p1, o1)+d (p2,o2)]≤ [d (p1, o2)+d (p2, o1)], the competition is
between p1 vs. o1 and p2 vs o2.
Otherwise, the competition is between p1 vs o2 and p2 vs o1.
Determine the winner. Individuals with higher fitness values win
the competition and will stay in the population; Losers are
discarded.
27
RTS works as follows
• Select two solutions at random, A and B, from the
population and perform crossover and mutation, resulting
in two new solutions, A´ and B´.
• For each new solution (A´ and B´), scan w individuals
(randomly chosen) from the population and pick the one
that is most similar to it. Call them A´´ and B´´.
• A´ competes with A´´. If A´ is better, then it replaces
A´´ in the population. (Do the same for B´and B´´)
– The algorithm is steady-state (solutions are introduced
in the population incrementally).
14
24-03-2014
RTS
• RTS performs much better than the standard crowding
mechanism.
• Also much simpler to use than fitness sharing.
• Harik used w as 4 times the number of desired peaks to be
found, and obtained good results.
Summary on niche EAs
15