+ All Categories
Home > Documents > Feasibility-Preserving Crossover for Maximum k -Coverage Problem

Feasibility-Preserving Crossover for Maximum k -Coverage Problem

Date post: 10-Jan-2016
Category:
Upload: jela
View: 38 times
Download: 0 times
Share this document with a friend
Description:
Feasibility-Preserving Crossover for Maximum k -Coverage Problem. Yourim Yoon Yong-Hyuk Kim Byung-Ro Moon Seoul National University. S 2. S 3. S 4. S 1. S 5. S 6. Covering Problems. Set covering problem # of elements = 12 # of subsets = 6 Min-size set cover? - PowerPoint PPT Presentation
Popular Tags:
22
Feasibility-Preserving Crossover for Maximum k- Coverage Problem Yourim Yoon Yong-Hyuk Kim Byung-Ro Moon Seoul National University
Transcript
Page 1: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Feasibility-Preserving Crossover for Maximum k-Coverage Problem

Yourim YoonYong-Hyuk KimByung-Ro Moon

Seoul National University

Page 2: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Covering Problems

Set covering problem # of elements = 12 # of subsets = 6 Min-size set cover? = {S2, S3, S4}

S2 S3 S4

S5

S6

S1

Page 3: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Maximum k-Coverage Problem

S2 S3 S4

S5

S6

S1

# of elements = 16 # of subsets = 6 k = 3 (# of used subsets) Maximum coverage with

3 subsets? = {S1, S3, S5}

(covers 13 elements)

Page 4: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Maximum k-Coverage Problem

A generalized version of covering problems Introduced by [Hochbaum and Pathria, 1998] NP-hard Many applications

Covering graphs by subgraphs Facility location problem Packing and circuit layout design Scheduling problems

Page 5: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Representation

Subsets {S1, S2, S3, S4}, n = 4, k = 2

Two encodings Binary encoding

Length-n binary string E.g., {S1, S3}

Integer encoding Length-k integer string of indices of selected subsets E.g., {S1, S3} or

Each phenotype is represented by k! genotypes

We consider integer encoding (since k << n)

1 0 1 0

1 3 3 1

Page 6: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Motivation

GA should produce offspring such that … They have as few common elements between genes (subsets)

as possible.

m = 5, n = 4, k = 2

1 2 3

4 5

S1

S2S3

S4

Solution Common

element set

Coverage

(size)

(1,2)

(1,3)

(1,4)

(2,3)

(2,4)

(3,4)

Empty

Empty

{2}

{5}

{3}

Empty

{1,2,3,5} (4)

{1,2,4,5} (4)

{1,2,3} (3)

{3,4,5} (3)

{2,3,5} (3)

{2,3,4,5} (4)

Page 7: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Motivation (cont’d)

Integer encoding is redundant Each phenotype is represented by k! genotypes

In crossover, Each gene (subset) of Parent 1 should match the gene

(subset) of Parent 2 with as many common elements as possible.

1 3 3 1=

Page 8: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

New Crossover

Let distance dij for each subset pair (Si, Sj) be given

Pseudo-code

STEP 1. Find an optimal assignment between genes of

Parent 1 and those of Parent 2;

STEP 2. Normalize Parent 2 using the above assignment;

STEP 3. Do traditional n-point crossover between Parent 1

and normalized Parent 2;

STEP 1 can be efficiently computed by Hungarian method.

Its time complexity is O(k3)

Page 9: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Optimal Assignment between Genes

Parent 1 Parent 2

Gene 1

Gene 2

Gene k

Gene 1

Gene 2

Gene k

Permutationσ

Minimize the summation of distances betweenGene i of Parent 1 and Gene σ(i) of Parent 2 for all i

Formally, min Σi diσ(i)

NormalizedParent 2

Gene σ(1)

Gene σ(2)

Gene σ(k)

Page 10: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Distances between Subsets

Hamming distance Distance between Si and Sj := |(Si−Sj)(Sj− Si)|

NH-Xover Normalized by Hamming distance

Discrete distance Distance between Si and Sj := I(Si = Sj)

ND-Xover Normalized by discrete distance

Page 11: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Distances between Subsets (cont’d)

An example Hamming distance

Discrete distancem = 5, n = 4, k = 2

S1

S2S3

S4Parent 1

Parent 2

1 3

2 3

1 2

Parent 1

Parent 2

1 3

2 3

1 2

Distance 4 + 0 = 4

Distance 1 + 0 = 1

Page 12: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

cut point

New Crossover (cont’d)

An example of NH-Xover

Parent 1

Parent 2

1 2

3 4

Distance 4 + 2 = 6

Parent 1

Normalized Parent 2

1 2

4 3

Distance 2 + 2 = 4

m = 5, n = 4, k = 2

S1

S2S3

S4

1 2

Distance between Si and Sj = |(Si−Sj) (Sj− Si)|

Page 13: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Traditional Approaches

Do not preserve feasibility Example. k = 4, n = 10

3 5 6 9

1 3 5 7

3 3 5 7

Parent 1

Parent 2

Offspring

0 0 1 0 1 1 0 0 1 0

1 0 1 0 1 0 1 0 0 0

0 0 1 0 1 1 0 0 0 0

Integer representation Binary representation

1 2 3 4 1 2 3 4 5 6 7 8 9 10

Page 14: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Useful Property

Theorem. Proposed crossover preserves feasibility

Parent 1

Parent 2

NormalizedParent 2

Page 15: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Experiments

Genetic Framework A variant of CHC [Eshelman, 1991]

Pseudo-codeMake an initial population (N individuals);do

Choose N/2 random pairs from population;Make N/2 offspring by crossover;Select the best N individuals for next population;if population has no change during T generation,

then reinitialize population;until (stop criterion); return the best solution found so far;

400 individuals

500 generations

Page 16: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Experiments (cont’d)

Test Instance Sets 11 classes, 65 instances from OR-library Instance sets for set covering problems

Instance set m n Density (%) # of instances

I-4

I-5

I-6

I-A

I-B

I-C

I-D

I-E

I-F

I-G

I-H

200

200

200

300

300

400

400

500

500

1,000

1,000

1,000

2,000

1,000

3,000

3,000

4,000

4,000

5,000

5,000

10,000

10,000

2

2

5

2

5

2

5

10

20

2

5

10

10

5

5

5

5

5

5

5

5

5

Page 17: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

0

0.5

1

1.5

2

2.5

3

3.5

4

I-4 I-5 I-6 I-A I-B I-C I-D I-E I-F I-G I-H

Aver

age

Instance set

%-g

ap NH-XoverND-Xover

Comparison between NH-Xover & ND-Xover

k = 10

Here, %-gap = 100 x |best − output| / best

Page 18: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

I-4 I-5 I-6 I-A I-B I-C I-D I-E I-F I-G I-H

Aver

age

Instance set

%-g

ap NH-XoverND-Xover

Comparison between NH-Xover & ND-Xover (cont’d)

k = 20

Here, %-gap = 100 x |best − output| / best

Page 19: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Conclusions

New crossover for maximum k-coverage problem Efficient running

Implemented using Hungarian method Preserving feasibility

Not necessary to repair

Future work Extensive empirical studies for various test set Comparison with traditional crossovers combined with

repairing On integer encoding & binary encoding

Page 20: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Thank you for listening!

Page 21: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Maximum k-Coverage Problem

m := # of elements, n := # of subsets

m = 5, n = 4

1 2 3

4 5

S1

S2S3

S4A = (aij) m x n 0-1 matrix

1 1 0 0 02 1 0 0 13 0 1 0 14 0 0 1 05 0 1 1 0( )

S1 S2 S3 S4

element

subset

Page 22: Feasibility-Preserving Crossover for Maximum  k -Coverage Problem

Maximum k-Coverage Problem

Formal definition xj = 1 iff the j th subset is selected

Here, I(true) = 1 & I(false) = 0

# of selected subsets = k

Weighted sum ofcovered elements


Recommended