Skip to main content

Advertisement

Log in

Ant colony optimization for triple product property triples to fast matrix multiplication

  • Methodologies and Application
  • Published:
Soft Computing Aims and scope Submit manuscript

Abstract

The group-theoretic approach to fast matrix multiplication has been introduced by Cohn and Umans (proceedings of 44th annual IEEE symposium on foundations of computer science, IEEE Computer Society, 2003). In this approach, there is a challenging problem. It is to find three subsets of a given group satisfying the so-called triple product property such that the product of their sizes is as large as possible. For this challenging problem, some brute-force algorithm has been proposed, which is exact but time-consuming. The ant colony optimization is a randomized heuristic algorithm and finds extensive applications in many fields. We use it to solve the problem of searching for three subsets of a given group such that they satisfy the triple product property and the product of their sizes reaches the maximum. Experimental results show that the ant colony optimization is efficient for this problem. Using this approach, we obtain three subsets for each nonabelian group of order not larger than 44 which satisfy the triple product property and the product of whose sizes reaches the best value found so far.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1

Similar content being viewed by others

Explore related subjects

Discover the latest articles, news and stories from top researchers in related subjects.

References

  • Ambainis A, Filmus Y, Le Gall F (2015) Fast matrix multiplication: limitations of the coppersmith-winograd method. In: Proceedings of the forty-seventh annual acm symposium on theory of computing, pp 585–593

  • Bini D, Capovani M, Romani F, Lotti G (1979) \(\text{ O }(n^{2.7799})\) complexity for \(n\times n\) approximate matrix multiplication. Inf Process Lett 8(5):234–235

    Article  MATH  Google Scholar 

  • Bshouty NH (1995) On the additive complexity of \(2\times 2\) matrix multiplication. Inf Process Lett 56(6):329–335

    Article  MATH  Google Scholar 

  • Cohn H, Umans C (2003) A group-theoretic approach to fast matrix multiplication. In: Proceedings of 44th annual IEEE symposium on foundations of computer science, IEEE Computer Society, pp 438–449

  • Coppersmith D, Winograd S (1982) On the asymptotic complexity of matrix multiplication. SIAM J Comput 11(3):472–492

    Article  MATH  MathSciNet  Google Scholar 

  • Coppersmith D, Winograd S (1990) Matrix multiplication via arithmetic progressions. J Symb Comput 9(3):251–280

    Article  MATH  MathSciNet  Google Scholar 

  • Dorigo M, Maniezzo V, Colorni A (1996) Ant system: optimization by a colony of cooperating agents. IEEE Trans Syst Man Cybern Part B Cybern 26(1):29–41

    Article  Google Scholar 

  • Hart S, Hedtke I, Müller-Hannemann M, Murthy S (2015) A fast search algorithm for \(<m, m, m>\) triple product property triples and an application for 5 \(\times \) 5 matrix multiplication. Groups Complexity Cryptol 7(1):31–46

    Article  MATH  MathSciNet  Google Scholar 

  • Hedtke I, Murthy S (2012) Search and test algorithms for triple product property triples. Groups Complex Cryptol 4(1):111–133

    Article  MATH  MathSciNet  Google Scholar 

  • Hopcroft JE, Kerr LR (1969) Some techniques for proving certain simple programs optimal. In: IEEE conference record of 10th annual symposium on switching and automata theory, pp 36–45

  • Jansen T, Wegener I (2001) Evolutionary algorithms: how to cope with plateaus of constant fitness and when to reject strings of the same fitness. IEEE Trans Evol Comput 5(6):589–599

    Article  Google Scholar 

  • Le Gall F (2014) Powers of tensors and fast matrix multiplication. Eprint Arxiv pp 296–303

  • Neumann F, Sudholt D, Witt C (2009) Analysis of different MMAS ACO algorithms on unimodal functions and plateaus. Swarm Intell 3(1):35–68

    Article  Google Scholar 

  • Neumann PM (2011) A note on the triple product property for subsets of finite groups. LMS J Comput Math 14(14):232–237

    Article  MATH  MathSciNet  Google Scholar 

  • Pan VY (1978) Strassen’s algorithm is not optimal, trilinear technique of aggregating, uniting and canceling for constructing fast algorithms for matrix operations. In: Proceedings of the 19th annual symposium on foundations of computer science, pp 166–176

  • Pan VY (1980) New fast algorithms for matrix operations. SIAM J Comput 9(2):321–342

    Article  MATH  MathSciNet  Google Scholar 

  • Rai P (2014) Edge detection of medical images using modified ant colony optimization algorithm based on weighted heuristics. Int J Image Gr Signal Process 6(3):21–26

    Article  Google Scholar 

  • Salama K, Abdelbar AM (2014) A novel ant colony algorithm for building neural network topologies. In: Proceedings of the 9th international conference on ant colony optimization and swarm intelligence, ANTS 2014, Brussels, Belgium, September 10–12, pp 1–12

  • Schönhage A (1981) Partial and total matrix multiplication. SIAM J Comput 10(3):434–455

    Article  MATH  MathSciNet  Google Scholar 

  • Stothers A (2010) On the complexity of matrix multiplication. Ph.D. thesis, University of Edinburgh

  • Strassen V (1969) Gaussian elimination is not optimal. Numer Math 13(4):354–356

    Article  MATH  MathSciNet  Google Scholar 

  • Strassen V (1987) Relative bilinear complexity and matrix multiplication. J Für Die Reine Und Angewandte Math 375–376:406–443

    MATH  MathSciNet  Google Scholar 

  • Stützle T, Hoos HH (2000) Max–Min ant system. Future Gener Comput Syst 16(8):889–914

    Article  MATH  Google Scholar 

  • Williams VV (2012) Multiplying matrices faster than coppersmith-winograd. In: Proceedings of the forty-fourth annual acm symposium on theory of computing, pp 887–898

  • Winograd S (1971) On multiplication of \(2\times 2\) matrices. Linear Algebra Appl 4(4):381–388

    Article  MATH  MathSciNet  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Yuren Zhou.

Ethics declarations

Funding

This study was funded by the National Natural Science Foundation of China (Nos. 61472143, 61562071), the Scientific Research Special Plan of Guangzhou Science and Technology Programme (No. 201607010045), and the Natural Science Foundation of Jiangxi Province (No. 20151BAB207020).

Conflict of interest

The authors declare that they have no conflict of interest

Human and Participants

This article does not contain any studies with human participants or animals performed by any of the authors.

Additional information

Communicated by V. Loia.

Appendix

Appendix

GAP Id

\(\langle n,m,p \rangle \)

Details subsets

[36, 1]

\(\langle 6, 4, 2\rangle \)

\(G:=\)SmallGroup(36,1), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_1*f_2*f_3,f_1*f_2*f_4,f_1*f_2*f_3^2\),

\(f_2*f_3*f_4^2,f_2*f_3^2*f_4^2]\);

\(T=[e,f_1,f_2,f_1*f_2]\);

\(U=[e,f_1*f_2*f_4^2]\);

[36, 3]

\(\langle 9\), 3, \(2\rangle \)

\(G:=\)SmallGroup(36,3), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_2,f_1*f_4,f_2^2,f_1^2*f_3,f_1*f_2*f_4\),

\(f_1^2*f_2*f_3,f_1*f_2^2*f_4,f_1^2*f_2^2*f_3]\);

\(T=[e,f_1,f_1^2*f_2]\);

\(U=[e,f_3*f_4]\);

[36, 4]

\(\langle \)12, 2\(, 2\rangle \)

\(G:=\)SmallGroup(36,4), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_3,f_1*f_3,f_2*f_3,f_1*f_3^2,f_1*f_4^2\),

\(f_2*f_4^2,f_3^2*f_4,f_1*f_2*f_3*f_4\),

\(f_1*f_2*f_4^2,f_2*f_3^2*f_4^2\),

\(f_1*f_2*f_3^2*f_4^2]\);

\(T=[e, f_1*f_3^2*f_4]\);

\(U=[e, f_1*f_3^2*f_4^2]\);

[36, 6]

\(\langle 12, 2, 2\rangle \)

\(G:=\)SmallGroup(36,6), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_3,f_2^2,f_1*f_2*f_4,f_1*f_4^2,f_2^2*f_3\),

\(f_2*f_4^2,f_1*f_2*f_3*f_4,f_1*f_3*f_4^2\),

\(f_2*f_3*f_4^2,f_1*f_2^2*f_4^2\),

\(f_1*f_2^2*f_3*f_4^2]\);

\(T=[e,f_1*f_4]\);

\(U=[e,f_1]\);

[36, 7]

\(\langle \)12, 2\(, 2\rangle \)

\(G:=\)SmallGroup(36,7), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_2,f_4,f_1*f_4,f_2*f_4,f_1*f_2*f_4\),

\(f_1*f_4^2,f_3^2*f_4,f_1*f_2*f_4^2,f_2*f_3^2*f_4\),

\(f_1*f_3^2*f_4^2,f_1*f_2*f_3^2*f_4^2]\);

\(T=[e,f_1*f_2*f_3^2]\);

\(U=[e,f_1*f_2*f_3*f_4^2]\);

[36, 9]

\(\langle \)12, 2\(, 2\rangle \)

\(G:=\)SmallGroup(36,9), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_1*f_3,f_2*f_3,f_3*f_4,f_4^2\),

\(f_1*f_3*f_4,f_1*f_2*f_3^2,f_1*f_3*f_4^2\),

\(f_2*f_3^2*f_4,f_2*f_3*f_4^2\),

\(f_1*f_2*f_3^2*f_4\),\(f_1*f_2*f_3^2*f_4^2]\);

\(T=[e,f_2]\);

\(U=[e,f_2*f_3^2]\);

[36, 10]

\(\langle 12, 2,2\rangle \)

\(G:=\)SmallGroup(36,10), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_1,f_3,f_1*f_3,f_2*f_4,f_1*f_2*f_4\),

\(f_1*f_4^2,f_2*f_3^2,f_3*f_4^2,f_1*f_2*f_4^2\),

\(f_2*f_3^2*f_4,f_1*f_2*f_3^2*f_4^2]\);

\(T=[e,f_1*f_2]\);

\(U=[e,f_1*f_2*f_3*f_4]\);

[36, 11]

\(\langle \)8, 2\(, 3\rangle \)

\(G:=\)SmallGroup(36,11), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_2,f_1*f_4,f_1^2*f_2,f_1^2*f_3*f_4\),

\(f_1*f_2^2*f_3,f_2^2*f_3*f_4\),

\(f_1^2*f_2^2*f_3*f_4]\);

\(T=[e,f_3*f_4]\);

\(U=[e,f_1*f_2*f_3*f_4,f_1^2*f_2^2*f_4]\);

[36, 12]

\(\langle 12, 2,2\rangle \)

\(G:=\)SmallGroup(36,12), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_1*f_2,f_3^2,f_1*f_3^2,f_1*f_3*f_4\),

\(f_1*f_4^2,f_2*f_3^2,f_2*f_3*f_4,f_2*f_4^2\),

\(f_3*f_4^2,f_1*f_2*f_3^2,f_1*f_2*f_3*f_4^2]\);

\(T=[e,f_1*f_2*f_4]\);

\(U=[e,f_1*f_2*f_4^2]\);

[36, 13]

\(\langle 12, 2,2\rangle \)

\(G:=\)SmallGroup(36,13), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_2,f_3,f_1*f_2,f_1*f_4,f_4^2\),

\(f_1*f_2*f_4,f_1*f_3^2,f_2*f_3*f_4\),

\(f_1*f_2*f_3*f_4,f_1*f_3*f_4^2\),

\(f_2*f_3^2*f_4^2]\);

\(T=[e,f_1*f_2*f_3]\);

\(U=[e,f_1*f_2*f_3^2*f_4^2]\);

[38, 1]

\(\langle 12,2,2\rangle \)

\(G:=\)SmallGroup(38,1), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2\)

\(S=[e,f_1*f_2^2,f_2^3,f_1*f_2^5,f_2^6,f_1*f_2^8\),

\(f_1*f_2^9,f_1*f_2^{10},f_2^{11},f_1*f_2^{13}\),

\(f_2^{14},f_1*f_2^{16}]\);

\(T=[e,f_1*f_2^{11}]\);

\(U=[e,f_1*f_2]\);

[39, 1]

\(\langle 6,3,3\rangle \)

\(G:=\)SmallGroup(39,1), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2\)

\(S=[e,f_2,f_1^2*f_2^2,f_2^9,f_1*f_2^{10},f_1^2*f_2^{11}]\);

\(T=[e,f_1^2*f_2^5,f_1*f_2^{11}]\);

\(U=[e,f_1*f_2^5,f_1^2*f_2^7]\);

[40, 1]

\(\langle 12,2,2\rangle \)

\(G:=\)SmallGroup(40,1), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_1,f_2,f_3,f_1*f_2,f_1*f_3,f_2*f_3\),

\(f_1*f_2*f_3,f_1*f_4^4,f_1*f_2*f_4^4\),

\(f_1*f_3*f_4^4,f_1*f_2*f_3*f_4^4]\);

\(T=[e,f_1*f_2*f_3*f_4^3]\);

\(U=[e,f_1*f_4]\);

[40, 3]

\(\langle 12,2,2\rangle \)

\(G:=\)SmallGroup(40,3), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_1,f_3,f_1*f_3,f_1*f_4,f_2*f_4\),

\(f_2*f_3*f_4,f_1*f_2*f_4^3,f_2*f_4^4\),

\(f_1*f_3*f_4,f_1*f_2*f_3*f_4^3\),

\(f_2*f_3*f_4^4]\);

\(T=[e,f_2*f_4^2]\);

\(U=[e,f_2*f_3*f_4^3]\);

[40, 4]

\(\langle 12,2,2\rangle \)

\(G:=\)SmallGroup(40,4), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_3,f_4^2,f_1*f_4^2,f_2*f_4^2,f_3*f_4^2\),

\(f_1*f_2*f_4^2,f_1*f_3*f_4^2,f_2*f_3*f_4^2\),

\(f_1*f_2*f_3*f_4^2,f_1*f_2*f_4^4\),

\(f_1*f_2*f_3*f_4^4]\);

\(T=[e,f_1*f_3*f_4^3]\);

\(U=[e,f_1*f_4^4]\);

[40, 5]

\(\langle 12,2,2\rangle \)

\(G:=\)SmallGroup(40,5), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_1*f_3,f_2*f_4^2,f_1*f_2*f_4^2,f_1*f_4^3\),

\(f_2*f_3*f_4^2,f_2*f_4^3,f_3*f_4^3\),

\(f_1*f_2*f_3*f_4^2,f_2*f_3*f_4^3,f_3*f_4^4\),

\(f_1*f_3*f_4^4 ]\);

\(T=[e,f_1*f_3*f_4^3]\);

\(U=[e,f_1*f_3*f_4]\);

[40, 6]

\(\langle 13,2,2\rangle \)

\(G:=\)SmallGroup(40,6), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_4,f_1*f_3,f_1*f_2*f_4,f_1*f_4^2\),

\(f_2*f_3*f_4,f_1*f_2*f_4^2,f_1*f_4^3\),

\(f_2*f_3*f_4^2,f_3*f_4^3,f_2*f_4^4\),

\(f_1*f_2*f_3*f_4^3,f_1*f_2*f_3*f_4^4]\);

\(T=[e,f_1*f_2*f_3]\);

\(U=[e,f_1*f_3*f_4^3]\);

[40, 7]

\(\langle 12,2,2\rangle \)

\(G:=\)SmallGroup(40,7), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_1,f_2,f_3,f_1*f_3,f_2*f_3,f_2*f_4^2\),

\(f_2*f_3*f_4^2,f_2*f_4^3,f_1*f_2*f_4^3\),

\(f_2*f_3*f_4^3,f_1*f_2*f_3*f_4^3]\);

\(T=[e,f_1*f_2*f_3*f_4]\);

\(U=[e,f_1*f_4^2]\);

[40, 8]

\(\langle 12,2,2\rangle \)

\(G:=\)SmallGroup(40,8), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_1*f_3,f_1*f_2*f_3,f_3*f_4^2,f_1*f_4^3\),

\(f_3*f_4^3,f_4^4,f_2*f_3*f_4^3,f_2*f_4^4\),

\(f_1*f_2*f_4^4,f_2*f_3*f_4^4\),

\(f_1*f_2*f_3*f_4^4]\);

\(T=[e,f_1*f_3*f_4^4]\);

\(U=[e,f_1*f_3*f_4^2]\);

[40, 10]

\(\langle 10,2,2\rangle \)

\(G:=\)SmallGroup(40,10), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_3,f_1*f_2*f_4,f_1*f_3^2,f_1*f_3*f_4\),

\(f_2*f_3^2*f_4,f_3^3*f_4,f_1*f_3^3*f_4\),

\(f_2*f_3^4,f_1*f_3^4*f_4]\);

\(T=[e,f_4]\);

\(U=[e,f_2]\);

[40, 11]

\(\langle 10,2,2\rangle \)

\(G:=\)SmallGroup(40,11), \(e=\)One(G), \(f_1=G.1\)

\(S=[e,f_4,f_1*f_2*f_3,f_2*f_3^2,f_3^3\),

\(f_1*f_2*f_3*f_4,f_2*f_3^2*f_4,f_3^4\),

\(f_3^3*f_4,f_3^4*f_4]\);

\(T=[e,f_1]\);

\(U=[e,f_2]\);

\(f_2=G.2,f_3=G.3,f_4=G.4\)

[40, 12]

\(\langle 12,2,2\rangle \)

\(G:=\)SmallGroup(40,12), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_2,f_3,f_1*f_4,f_2*f_4,f_1*f_3*f_4\),

\(f_2*f_3*f_4,f_1*f_2*f_3*f_4,f_4^4,f_1*f_4^4\),

\(f_1*f_2*f_4^4,f_1*f_2*f_3*f_4^4]\);

\(T=[e,f_3*f_4^4]\);

\(U=[e,f_3*f_4^2]\);

[40, 13]

\(\langle 12,2,2\rangle \)

\(G:=\)SmallGroup(40,13), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3,f_4=G.4\)

\(S=[e,f_2,f_1*f_3,f_3*f_4,f_1*f_2*f_3\),

\(f_1*f_4^2,f_2*f_3*f_4\),\(f_4^3,f_1*f_2*f_4^2\),

\(f_2*f_4^3,f_3*f_4^3,f_1*f_2*f_3*f_4^3]\);

\(T=[e,f_1*f_2*f_4]\);

\(U=[e,f_1]\);

[42, 1]

\(\langle 12,2,2\rangle \)

\(G:=\)SmallGroup(42,1), \(e=\)One(G), \(f_1=G.1\)

\(f_2=G.2,f_3=G.3\)

\(S=[e,f_3,f_2^2,f_3^2,f_1*f_3^2,f_2*f_3^2\),

\(f_1*f_2*f_3^2,f_2^2*f_3^3,f_2^2*f_3^4,f_2*f_3^5\),

\(f_1*f_2^2*f_3^4,f_1*f_2*f_3^6]\);

\(T=[e,f_1*f_3^6]\);

\(U=[e,f_1*f_3^3]\);

[42, 2]

\(\langle 6,3,3\rangle \)

\(G:=\)SmallGroup(42,2), \(e=\)One(G),

\(f_1=G.1,f_2=G.2,f_3=G.3\)

\(S=[e,f_2*f_3,f_2^2*f_3^3,f_1*f_2*f_3^4\),

\(f_1*f_2*f_3^5\), \(f_1*f_2^2*f_3^6]\);

\(T=[e,f_2^2*f_3,f_2*f_3^5]\);

\(U=[e,f_2^2*f_3^2,f_2*f_3^3]\);

[42, 3]

\(\langle 14,2,2\rangle \)

\(G:=\)SmallGroup(42,3), \(e=\)One(G),

\(f_1=G.1,f_2=G.2,f_3=G.3\)

\(S=[e,f_1*f_2,f_1*f_3,f_2^2,f_2*f_3^2\),

\(f_1*f_2^3,f_1*f_2^2*f_3,f_2^4*f_3,f_2^3*f_3^2\),

\(f_2^6,f_2^5*f_3,f_1*f_2^4*f_3^2\),

\(f_1*f_2^6*f_3,f_1*f_2^5*f_3^2]\);

\(T=[e,f_1]\);

\(U=[e,f_1*f_3^2]\);

[42, 4]

\(\langle 12,2,2\rangle \)

\(G:=\)SmallGroup(42,4), \(e=\)One(G),

\(f_1=G.1,f_2=G.2,f_3=G.3\)

\(S=[e,f_1,f_1*f_2,f_1*f_3,f_2^2*f_3^2,f_2*f_3^3\),

\(f_1*f_2*f_3^3,f_1*f_3^4,f_1*f_2^2*f_3^3\),

\(f_1*f_2*f_3^4,f_2^2*f_3^6,f_1*f_2^2*f_3^6]\);

\(T=[e,f_1*f_3^5]\);

\(U=[e,f_1*f_3^3]\);

[42, 5]

\(\langle 14,2,2\rangle \)

\(G:=\)SmallGroup(42,5), \(e=\)One(G),

\(f_1=G.1,f_2=G.2,f_3=G.3\)

\(S=[e,f_1*f_2,f_2*f_3,f_3^2,f_1*f_2^2*f_3\),

\(f_1*f_2*f_3^2,f_1*f_3^3,f_2^2*f_3^3,f_2*f_3^4\),

\(f_1*f_2*f_3^4,f_1*f_3^5,f_2^2*f_3^5,f_2*f_3^6\),

\(f_1*f_2^2*f_3^6]\);

\(T=[e,f_1*f_2*f_3^3]\);

\(U=[e,f_1*f_2^2*f_3^3]\);

[44, 1]

\(\langle 11,2,2\rangle \)

\(G:=\)SmallGroup(44,1), \(e=\)One(G),

\(f_1=G.1,f_2=G.2,f_3=G.3\)

\(S=[e,f_3,f_2*f_3^3,f_1*f_2*f_3^3,f_3^5,f_1*f_3^5\),

\(f_1*f_2*f_3^7,f_1*f_3^8,f_1*f_2*f_3^9\),

\(f_2*f_3^{10},f_1*f_2*f_3^{10}]\);

\(T=[e,f_2]\);

\(U=[e,f_1*f_2*f_3]\);

[44, 3]

\(\langle 14,2,2\rangle \)

\(G:=\)SmallGroup(44,3), \(e=\)One(G),

\(f_1=G.1,f_2=G.2,f_3=G.3\)

\(S=[e,f_1*f_3,f_3^2,f_2*f_3^2,f_1*f_3^4\),

\(f_3^5,f_1*f_2*f_3^4,f_1*f_3^6,f_1*f_3^7\),

\(f_2*f_3^7,f_2*f_3^8,f_3^{10},f_1*f_2*f_3^9\),

\(f_1*f_2*f_3^{10}]\);

\(T=[e,f_1]\);

\(U=[e,f_1*f_2*f_3]\);

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Lai, X., Zhou, Y. & Xiang, Y. Ant colony optimization for triple product property triples to fast matrix multiplication. Soft Comput 21, 7159–7171 (2017). https://doi.org/10.1007/s00500-016-2259-y

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00500-016-2259-y

Keywords

Navigation