On Erdős conjecture for multiplicities

Antoine Deza, Frantisek Franek, and Min Jing Liu
Advanced Optimization Laboratory
Department of Computing and Software
McMaster University, Hamilton, Ontario, Canada


Table 1 The combination case used for produce 7-cliques

i

combination of set

1

{7}

2

{1,6} {2,5} {3, 4}

3

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

4

{1,1,1,4}{1,1,2,3}{1,2,2,2}

5

{1,1,1,1,3}{1,1,1,2,2}

6

{1,1,1,1,1,2}

7

{1,1,1,1,1,1,1}

Table 2 The coefficient of Si(X; F).

t

s1

s2

s3

s4

s5

s6

s7

4

7

6

1

 

 

 

 

5

15

25

10

1

 

 

 

6

31

90

65

15

1

 

 

7

63

301

350

140

21

1

 

8

127

966

1701

1050

266

28

1

Table 3 The coefficient of ki(X; F)

t

k1

k2

k3

k4

k5

k6

k7

k8

4

1

14

36

24

 

 

 

 

5

1

30

150

240

120

 

 

 

6

1

62

540

1560

1800

720

 

 

7

1

126

1806

8400

16800

15120

5040

 

8

1

254

5796

40824

126000

191520

141120

40320

 

 The relationship between x_i and corresponding coefficient in S_i

 0 means “>”, and 1 means “=”; eg  x0 > x1 = x2 > x3 then “010”

S_2

 

 

S_5

 

 

 coe1

0

2

coe4

0000

120

 

1

1

 

0001

60

S_3

 

 

 

0010

60

coe2

00

6

 

0011

20

 

01

3

 

0100

60

 

10

3

 

0101

30

 

11

1

 

0110

20

S_4

 

 

 

0111

5

coe3

000

24

 

1000

60

 

001

12

 

1001

30

 

010

12

 

1010

30

 

011

4

 

1011

10

 

100

12

 

1100

20

 

101

6

 

1101

10

 

110

4

 

1110

5

 

111

1

 

1111

1

 

 

Testing Result:

Table 1: Using a full family, The number of i - tuples by using two approaches:

S_i is the number of i-tuples by using the newest approaches; s_i(old) is the number of i-tuples by using the old approaches[1].

j

3

4

5

6

S1

7

15

31

63

S1(old)

7

15

31

63

S2

42

210

930

3,906

S2(old)

42

210

930

3,906

S3

210

2,730

26,970

238,266

S3(old)

210

2,730

26,970

238,266

S4

840

32,760

755,160

14,295,960

S4(old)

840

32,760

755,160

14,295,960

S5

2520

360,360

20,389,320

843,461,640

S5(old)

2520

360,360

20,389,320

843,461,640

S6

5040

3,603,600

530,122,320

48,920,775,120

S6(old)

5040

3,603,600

530,122,320

48,920,775,120

S7

5040

32,432,400

13,253,058,000

2,788,484,181,840

S7(old)

5040

32,432,400

13,253,058,000

2,788,484,181,840

 

c5 = 0.885833698 * 2 ^ (-9); @(10,{1,3,4,7,8,10})

 

1

2

3

4

S_T

517

134970

17273850

1009617840

S’_T

506

125730

14734170

742203000

c6 = 0.744442503 * 2 ^(-14) @ (10,{1,3,4,7,8})

 

1

2

3

4

5

S_T

505

125010

14562090

726780600

13191935400

S’_T

518

135726

17463606

1028265840

26106252480

c7 =0.686897717 * 2 ^ (-20) @(11,{3,4,7,8,10,11})

 

1

2

3

4

5

6

S_T

1002

490050

113148090

11590147800

5.06501E+11

1.46774E+13

S’_T

1045

556842

146860362

17896958640

9.50437E+11

2.13599E+13

c8 =0.70014091 * 2 ^ (-27) @ (12,{1,3,4,7,8,11,12}) (No blue search) (Link to the select testing seed)

 

[1] Franek, F.: A note on Erdős conjecture on multiplicities of complete subgraphs { lower upper bound for cliques of size 6. Combinatorica, No. 3, Vol. 22, 451-454 (2002)