1. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Structural correlations and critical phenomena of random scale-free networks: Analytic approaches DOOCHUL KIM (Seoul National University) Collaborators: Byungnam Kahng (SNU), Kwang-Il Goh (SNU/Notre Dame), Deok-Sun Lee (Saarlandes), Jae- Sung Lee (SNU), G. J. Rodgers (Brunel) D.H. Kim (SNU)

2. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Outline I.Static model of scale-free networks II.Vertex correlation functions III.Number of self-avoiding walks and circuits IV.Percolation transition V.Critical phenomena of spin models defined on the static model VI.Conclusion

3. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) I.Static model of scale-free networks static model

4. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) static model  We consider sparse, undirected, non-degenerate graphs only. 1.Degree of a vertex i : 2.Degree distribution: = adjacency matrix element (0,1)

5. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) static model  Grandcanonical ensemble of graphs:  Static model: Goh et al PRL (2001), Lee et al NPB (2004), Pramana (2005, Statpys 22 proceedings), DH Kim et al PRE(2005 to appear) Precursor of the “hidden variable” model [Caldarelli et al PRL (2002), Soederberg PRE (2002), Boguna and Pastor-Satorras PRE (2003)]

6. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) static model 1.Each site is given a weight (“fitness”) 2.In each unit time, select one vertex i with prob. P i and another vertex j with prob. P j. 3.If i=j or a ij =1 already, do nothing (fermionic constraint). Otherwise add a link, i.e., set a ij =1. 4.Repeat steps 2,3 NK times ( K = time = fugacity =  L  /N ).  Construction of the static model m = Zipf exponent When m =0  ER case. Walker algorithm (+Robin Hood method) constructs networks in time O(N).  N=10 7 network in 1 min on a PC. Comments

7. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) static model  Such algorithm realizes a “grandcanonical ensemble” of graphs G ={ a ij } with weights   Each link is attached independently but with inhomegeous probability f i,j.

8. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) static model Degree distribution

9. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) static model Recall When l>3 ( 0<m<1/2 ), When 2<l<3 ( 1/2<m<1 ) f ij  Comments 0 1 1 3-l3-l 3-l3-l f ij  2KNP i P j f ij  1 Strictly uncorrelated in links, but vertex correlation enters (for finite N) when 2< l <3 due to the “fermionic constraint” (no self- loops and no multiple edges).

10. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) II. Vertex correlation functions Vertex correlations Related work: Catanzaro and Pastor-Satoras, EPJ (2005)

11. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Vertex correlations Simulation results of k nn (k) and C(k)

12. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Vertex correlations Our method of analytical evaluations: For a monotone decreasing function F(x), Use this to approximate the first sum as Similarly for the second sum.

13. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Vertex correlations Result (1) for

14. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Vertex correlations Result (2) for

15. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) N 2 1010 3 4 5 6 7 8 9 l Result (3) Finite size effect of the clustering coefficient for 2< l <3: Vertex correlations

16. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) III.Number of self-avoiding walks and circuits Number of SAWs and circuits The number of self-avoiding walks and circuits (self-avoiding loops) are of basic interest in graph theory. Some related works are: Bianconi and Capocci, PRL (2003), Herrero, cond-mat (2004), Bianconi and Marsili, cond-mat (2005) etc. Issue: How does the vertex correlation work on the statistics for 2< l <3 ?

17. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Number of SAWs and Circuits The number of L-step self-avoiding walks on a graph is where the sum is over distinct ordered set of (L+1) vertices, We consider finite L only. The number of circuits or self-avoiding loops of size L on a graph is with the first and the last nodes coinciding.

18. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Number of SAWs and Circuits Strategy for 2< l <3: For upper bounds, we use repeatedly. Similarly for lower bounds with The leading powers of N in both bounds are the same. Note: The “surface terms” are of the same order as the “bulk terms”. and

19. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Number of SAWs and Circuits Result(3): Number of L-step self-avoiding walks For, straightforward in the static model For, the leading order terms in N are obtained.

20. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Number of SAWs and Circuits Typical configurations of SAWs (2<λ<3) L=2 L=3 L=4 H HH SS BB BB BB SS

21. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Number of SAWs and Circuits Results(4): Number of circuits of size

22. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) IV. Percolation transition Percolation transition Lee, Goh, Kahng and Kim, NPB (2004)

23. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Percolation transition The static model graph weight can be represented by a Potts Hamiltonian,

24. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Percolation transition Partition function: ’

25. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Thermodynamic limit Exact analytic evaluation of the Potts free energy: Explicit evaluations of thermodynamic quantities. 1.Vector spin representation  2.Integral representation of the partition function 3.Saddle-point analysis l>4l>4 l=4.8 3<l<43<l<4 l=3. 6 2<l<32<l<3 l=2. 4 Percolation transition at Percolation transition

26. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) V.Critical phenomena of spin models defined on the static model Spin models on SM

27. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Spin models on SM Spin models defined on the static model network can be analyzed by the replica method.

28. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Spin models on SM The effective Hamiltonian reduces to a mean-field type one with infinite number of order parameters, When J i,j are also quenched random variables, extra averages on each J i,j should be done.

29. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Spin models on SM We applied this formalism to the Ising spin-glass [DH Kim et al PRE (2005)] Phase diagrams in T-r plane for l > 3 and l <3

30. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Spin models on SM Critical behavior of the spin-glass order parameter in the replica symmetric solution: To be compared with the ferromagnetic behavior for 2< l <3;

31. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) VI. Conclusion 1.The static model of scale-free network allows detailed analytical calculation of various graph properties and free-energy of statistical models defined on such network. 2.The constraint that there is no self-loops and multiple links introduces local vertex correlations when l, the degree exponent, is less than 3. 3.Two node and three node correlation functions, and the number of self-avoiding walks and circuits are obtained for 2< l <3. The walk statistics depend on the even-odd parity. 4.Kasteleyn construction of the Potts model is utilized to calculate thermodynamic quantities related to the percolation transition such as the mean number of independent loops. 5.The replica method is used to obtain the critical behavior of the spin-glass order parameters in the replica symmetry solution.

32. ICTP School and Workshop on Structure and Function of complex Networks (16-28 May 2005) Static Model N=3 1 Efficient method for selecting intergers 1, 2, , N with probabilities P 1, P 2, , P N. Walker algorithm ( )