Download presentation
Presentation is loading. Please wait.
Published byDwayne Owen Modified over 9 years ago
1
evolution the only process with a track-record of developing intelligent minds the process... “blind” & long timescale but... robust, adaptable, minimal requirements
2
natural evol – timescale (1) 4000 MYA self-replicating RNA 3500 MYA bacteria & proto-photosynthesis 3000 MYA cyanobacteria “pollute” atmosphere with oxygen 1200 MYA sexual reproduction arises 1000 MYA multi-cellular organisms 505 MYA vertebrates 475-450 MYA plants & arthropods on land 300ish MYA insects diversify, land-based reptiles, ferns 150 MYA proto-birds 125 MYA small modern mammal 110 MYA dinosaurs
3
natural evol – timescale (1) 65 MYA mass extinction no.5 (dinosaur extinction) primate order starts 30 & 15 MYA primates diverge 700 KYA common ancestor humans & Neanderthals 150 KYA mitochondrial eve (90-200kya) 100 KYA modern humans 50 KYA human expansion 35 KYA symbolism in humans 3 KYA iron tools AD 0001human population 150 million AD 1853human population 1 billion AD 2010human population exceeds 7 billion
4
natural evolution fundamentals genomes & phenomes parental selection mutation & recombination
5
humans from genes DNA (4 bases ATCG, a digital code) 23 x 2 chromosomes (volumes of design) DNA → amino acid → protein → organ → human body body construction complex, many gene interactions
6
artificial evolution genomes & phenomes (pipe bends, aerofoils & software) parent selection mutation & recombination timescales a “blind & selfish” process
7
AI evolution Alife population evolution other genetic algorithms
8
population evolution an example
9
a sample world
10
genomes & phenomes binary genome rule based phenomes rules specified using sensors & effectors
12
converting ‘genes’ to rules
13
crossover & mutation
14
encoded rules (previous example)
15
genome modification cross-over & mutation rate parameters –consider max/minimal mutation –consider max/minimal cross-over evolving parameter values
16
the process Create initial population Convert individuals to rulesets Test & score rulesets Select parents & generate new population
17
parent selection (basic example) IDFsum-F fred20 (20+0) sue1535(15+20) bill1954(19+35) sally3488(34+54) nancy391(3+88) steve9100(9+91)...etc... if random( 100 ) => 76 76 > 54 and 76 <= 88, 76 is in Sally’s sum-F bucket so Sally chosen
18
selection strategies elitism weak & strong selection population toroids –clusters –wandering mates puberty & old-age
19
toriod
20
toriod clusters
21
toriod - wandering
22
testing single tests & sets of tests static tests & progressive tests evolving tests solution & test - predator / prey predator/prey with toriods, etc
23
peformance profiles average performance variance graphs borrowed from Peter Andras, Newcastle University
24
rule interpretation 1.Rules: {ante => conseq} use First Firing 2.Rules: {ante => conseq / weight} use max weight NB: may also use certainty factors or fuzzy logic 3.Rules : {ante => effector / weight} use effector with max accumulated weight NB: may also use certainty factors or fuzzy logic 4.Rule networks… {N: ante => conseq || on-fail-goto} 5.{N: ante => conseq, weight, threshold => on-success-goto || on-fail-goto} 6.activity nets { f( g(a,b), h(c,d)) / Г } …etc
25
summary #1 you need well constructed... problem encoding (world, sensors, effectors, fn.s, etc) NB: if you don't know what is/isn't important then over supply phenome construction (avoid ambiguity) evolutionary operators (xover, mutate...) test strategy & fitness evaluation (be prepared to "tune" fitness function to improve performance) mate selection
26
summary #2 elitism, over-strong selection => quick elimination of "weaker" performers. May also remove useful but "sleeping" genes from population. => early convergence
27
summary #3 narrow spread of performances => convergence (ie: little genetic variation) large spread of the performances check: problem encoding, genetic operators & selection strategy ditto for too fast/slow increase in average performance
28
try questions from page 24 of... http://www.staff.ncl.ac.uk/peter.andras/evocompapp l.ppt
29
tutorial example
30
details 60 bit genome 20 * 3 bit N/S/W/E moves (+ 4 n/a moves) tester judges progression to SE corner typically sensitive to... population size selection strategy etc
31
in lisp... (defun evolve () (report-header) (create-initial-population) (dotimes (n *no-generations*) (generate-phenomes) (tester) (report n) (when (last-generation...)...) (setf *population* (breeder)) ))
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.