G??sse?%20pe?????s - PowerPoint PPT Presentation

About This Presentation
Title:

G??sse?%20pe?????s

Description:

St???e??d?? pe?????s ?? : s??se?? pe?????s ?? e ???s ata ... thrashing. redundant work. late detection of conflict. Systematic Search Methods. A. C ... – PowerPoint PPT presentation

Number of Views:32
Avg rating:3.0/5.0
Slides: 31
Provided by: scie381
Category:
Tags: 20pe | sse | thrashing

less

Transcript and Presenter's Notes

Title: G??sse?%20pe?????s


1
G??sse? pe?????sµ??
  • ?? e??a? ?? pe?????sµ??(constraints)?
  • ???te??p???s? pe?????sµ??
  • ?p???s? pe?????sµ??
  • ???d?a pe?????sµ??
  • ???e? pe?????? pe?????sµ??
  • Properties of constraint solving

2
?e?????sµ??
?etaß??t?? ap????e?s? t?µ??
S??a?t?s?a?? s?µß??a a?t?st????s? t?µ?? se t?µ??
S?es?a?? s?µß??a s??se?? µeta?? t?µ??
3
?e?????sµ??
St???e??d?? pe?????sµ?? s??se?? pe?????sµ?? µe
???sµata
unary
binary
?e?????sµ?? s?????s? st???e??d?? pe?????sµ??
4
??a??p???s?µ?t?ta
?p?t?µ?s? a???es? t?µ?? st?? µetaß??t??
?p???s? ap?t?µ?s? p?? ??a??p??e? t???
pe?????sµ???
5
??a??p???s?µ?t?ta
??a??p???s?µ? ? pe?????sµ?? ??e? ??s? ??
??a??p???s?µ? ? pe?????sµ?? de? ??e? ??s?
??a??p???s?µ?
?? ??a??p???s?µ?
6
S??ta?? pe?????sµ??
  • ?? pe?????sµ?? e??a? s?µß???se????
  • ?? a????e? de? ep??e????? (de? t??
    ???s?µ?p????µe)
  • ? se??? ep??e??e?
  • ?e????? a??????µ?? e?a?t??ta? ap? t?? se???

7
?s?d??aµ?a pe?????sµ??
??? d?af??et???? pe?????sµ?? µp??e? ?a
a?apa??st??? t?? ?d?a p????f???a
??? pe?????µ?? e??a? ?s?d??aµ?? ?ta? ????? t?
?d?? s????? ??se??
8
???te??p???s? ???s?µ?p????ta? pe?????sµ???
  • ?? pe?????sµ?? pe?????f??? ?deat?? s?µpe??f????
    a?t??e?µ???? st?? p?a?µat??? ??sµ?

9
??a??p???s? pe?????sµ??
  • ?????t?? e??? pe?????sµ?? C ????µe d?? e??t?se??
  • ??a??p???s?µ? ??e? ??s?
  • ep???s? d?se µ?? µ?a ??s?, e?? ?p???e?
  • ?? p??t? e??a? p?? ßas???
  • ??a s?st?µa ep???s?? pe?????sµ?? apa?t?e? t?
    e??t?µa t?? ??a??p???s?µ?t?ta?

10
??a??p???s? pe?????sµ??
  • ??? apa?t?µe t?? e??t?s?
  • ?p?? p??s????s? ap?t?µ?s? ???? t?? p???a???
    t?µ??(genarate test)

11
Generate Test
  • ???a??? ? p?? ?e???? µ???d?? ep???s??
  • ???????µ??
  • generate labelling
  • test ??a??p???s?µ?t?ta?
  • ?e???e?t?µata ?e?t??se??
  • blind generator smart generator --gt local
    search
  • a??? e?a???? ??e???? ?at? t?? e?a????
  • as??epe??? --gt backtracking

12
Backtracking (BT)
Systematic Search Methods
  • incrementally extends a partial solution towards
    a complete solution
  • Algorithm
  • assign value to variable
  • check consistency
  • until all variables labelled
  • Drawbacks
  • thrashing
  • redundant work
  • late detection of conflict

A
1
B
2
1
C
2
1
1
D
2
2
1
1
1
?
A D, B ? D, AC lt 4
13
GT BT - Example
Systematic Search Methods
  • Problem X1,2, Y1,2, Z1,2 X Y, X
    ? Z, Y gt Z
  • generate test backtracking

14
Consistency Techniques
  • removing inconsistent values from variables
    domains
  • graph representation of the CSP
  • binary and unary constraints only (no problem!)
  • nodes variables
  • edges constraints
  • node consistency (NC)
  • arc consistency (AC)
  • path consistency (PC)
  • (strong) k-consistency

Agt5
A
AltC
A?B
C
B
BC
15
Arc Consistency (AC)
Consistency Techniques
  • the most widely used consistency technique (good
    simplification/performance ratio)
  • deals with individual binary constraints
  • repeated revisions of arcs
  • AC-3, AC-4, Directional AC

a b c
a b c
a b c
Y
X
Z
16
AC - Example
Consistency Techniques
  • Problem X1,2, Y1,2, Z1,2 X Y, X
    ? Z, Y gt Z

X
X
1 2
1 2
1 2
1 2
Y
Y
1 2
1 2
Z
Z
17
Is AC enough?
Consistency Techniques
  • empty domain gt no solution
  • cardinality of all domains is 1 gt solution
  • Problem X1,2, Y1,2, Z1,2 X ? Y, X
    ? Z, Y ? Z

X
1 2
1 2
Y
Z
1 2
18
Path Consistency (PC)
Consistency Techniques
V2
  • consistency along the path only
  • checking paths of length 2 is enough
  • Plus/Minus
  • detects more inconsistencies than AC
  • - extensional representation of constraints
  • - changes in graph connectivity
  • Directional PC, Restricted PC

V4
V3
V5
V0
V1
???
19
K -consistency
Consistency Techniques
  • K-consistency
  • consistent valuation o (K-1) variables can be
    extended to K-th variable
  • strong K-consistency ? J-consistency for each
    J?K
  • NC ? strong 1-consistency
  • AC ? strong 2-consistency
  • PC ? strong 3-consistency

20
??a??p???s? pe?????sµ??
  • ? ap?? p??s????s? de? d???e?e? ??a p??ß??µata t??
    p?a?µat???? ??sµ??
  • ??a p?? ???p?? ??d?s? ???s?µ?p??e?ta? ??a
    pe?????sµ??? µe pepe?asµ??? ped?? ???sµ??
  • Gauss-Jordan s??µa apa???f??

21
Gauss-Jordan s??µa apa???f??
  • ?p????? e??s?s?? c ap? C
  • ?a?a????e c st? µ??f? x e
  • ??t??at?st?se t? x pa?t?? st? C µe e
  • ?pa???aße µ????
  • ??e? ?? e??s?se?? ?a e??a? t?? µ??f?? x e
  • ? µ?a e??s?s? e??a? ?s?d??aµ? µe d 0 (d ! 0)
  • ?p?st?e?e True st?? p??t? pe??pt?s? a????? False

22
Gauss-Jordan ?a??de??µa 1
??t??at?st?se X µe 2YZ-1
??t??at?st?se Y µe -1
?p?st?e?e false
23
Gauss-Jordan Example 2
??t??at?st?se X µe 2YZ-1
??t??at?st?se Y µe -1
?p???s? µ??f?? pe?????sµ?? a?t?? t?? µ??f??
e??a? ??a??p???s?µ??
24
?p???s? µ??f??
  • ?? pa?aµet????? µetaß??t?? eµfa?????ta? sta
    a??ste?? µ?a? e??s?s??.
  • ?a?aµet????? µetaß??t?? eµfa?????ta? sta de???
    µ?a? e??s?s??.
  • ?p???s? ep????? pa?aµe????? t?µ?? ?a? ?a????sµ??
    t?? µ? pa?aµet?????

25
?????
  • ?e???t? ???µ?s???
  • Scene Labelling
  • ??????? ?????aµµat?sµ??
  • unification --gt constraint solving
  • G?af???? a????ep?d??se??
  • Sketchpad
  • ThingLab
  • ?p??e???s?a?? ??e??a
  • NP-hard combinatorial problems

26
?fa?µ????
  • ???ß??µata ????es??
  • t?p??es?a ?atas?e??? ae??d??µ???
  • a?????ß???? p?????
  • p??s?p??? a???es?
  • ?at?????? ??s???µ??
  • a???es? p????µ?t?? pt?se??
  • d?a?e???s? ?a? d?aµ??f?s? d??t???
  • s?ed?asµ?? ?a??d??s?? se t??ep????????a?? d??t?a
  • ß??t?st? t?p???t?ts? sta?µ?? se as??µata d??t?a
  • µ???a?? ß??????a
  • DNA a???????a?
  • a?a??????? ?a? ??f?a??? s?ed?asµ?? ?????µ?t??

27
?fa?µ???? se ?????d?a???µµata
  • ? p?? ep?t???? pe????? efa?µ????
  • ?????d????aµµa pa?a?????
  • s?ed?asµ?? pa?a????? ae??p????? (Dassault
    Aviation)

28
?e???e?t?µata
  • NP-hard p??ß??µata
  • µ? p??ß????µ? s?µpe??f???
  • sta?e??t?ta µ??te??p???s??
  • p??? ???? ep?ped? (???? pe?????sµ??, ep???t??)
  • p??? ?aµ??? ep?ped? (µ??te??p???s?)
  • p??? t?p???
  • µ? a???t???? (rescheduling)
  • ad??aµ?a s??e??as?a? ep???t??

29
??e??e?t?µata
  • epe????µat??? f?s?
  • ep????t??s? t?? e?d?af???? st? pe????af? t??
    p??ß??µat?? ?a? ??? st?? ep???s? t??
  • ap?te?esµat??? ep???s? p??ß??µ?t??
  • e??p???s? a??????µ?? p?? ???s?µ?p?????ta? ??a
    e?d????? s??p???
  • ?eµe???d?? s?µas??????a
  • ?a?a??? ?a? ??µ??? ???sse?
  • ßas??eta? st? ?????? p????aµµat?sµ?
  • p??????e? efa?µ????
  • a??et? ep?t???a se d??f????? t?µe??

30
?ate????se??
  • µ??te??p???s?
  • ?a???????? pe?????sµ??? (???? d?af??et????)
  • µ??te??p???s? ???ss?? (Numerica, VisOpt)
  • ?ata???t? µe??t?
  • ?pt???p???s?, ap?te?esµat??? debugging
  • ?ß??d???? a??????µ??
  • s??e??as?a ep???t??
  • pa?a?????sµ??
  • multi-agent te???????a
Write a Comment
User Comments (0)
About PowerShow.com