Research Article | Open Access
Yilun Shang, "Phase Transition in Long-Range Percolation on Bipartite Hierarchical Lattices", The Scientific World Journal, vol. 2013, Article ID 172393, 5 pages, 2013. https://doi.org/10.1155/2013/172393
Phase Transition in Long-Range Percolation on Bipartite Hierarchical Lattices
We propose a family of bipartite hierarchical lattice of order N governed by a pair of parameters and . We study long-range percolation on the bipartite hierarchical lattice where any edge (running between vertices of unlike bipartition sets) of length k is present with probability , independently of all other edges. The parameter is the percolation parameter, while describes the long-range nature of the model. The model exhibits a nontrivial phase transition in the sense that a critical value if and only if and . Moreover, the infinite component is unique when .
For an integer , the hierarchical lattice of order is defined by The hierarchical distance on is defined by which satisfies the strong (non-Archimedean) triangle inequality: for any . This means that is an ultrametric space. Roughly speaking, this corresponds to the leaves of an infinite -ary tree, with metric distance half the graph distance.
Some stochastic models based on hierarchical lattices have been studied. The asymptotic long-range percolation on is analyzed in  for . To the best of our knowledge, this is the first paper devoted to . For different purpose, the works [2–4] study the long-range percolation on for fixed by using different connection probabilities. The contact process and perturbation analysis on for finite have been studied in [5, 6], respectively. Random walks on hierarchical lattices have been examined in [7, 8].
In this paper, we study percolation on a class of bipartite hierarchical lattices, where edges always run between vertices of unlike type. Bipartite graphs have been studied intensively in the literature (see e.g., [9, 10]) and bipartite structure is popular in many social networks including sexual-contact networks  and affiliation networks , but we have not seen the setup that we consider here. For two integers and , consider a partition of into two sets: Vertices in and are said to have types 1 and 2, respectively. For each , the probability of connection between two vertices and of unlike type such that is given by where and , all connections being independent. Vertices of the same type cannot be connected with each other, and hence the resulting graph is a class of random bipartite graph.
In the above bipartite hierarchical lattice, denoted by , vertices of both types are countable and the shortest distance between vertices in and is . The vertices in can be represented by the leaves at the bottom of an infinite regular tree, where branches emerge from each inner node, see Figure 1. The distance between two vertices (leaves at level 0) is the number of levels from the bottom to their most recent common ancestor. The partition of types for vertices is determined by their ancestors at level ; in other words, we need to track back at least levels to find the most recent common ancestor of two vertices of unlike type.
Two vertices are in the same component if there exists a finite sequence such that each pair of vertices and has different types and shares an edge for . In our model, the parameter describes the long-range nature, while we think of as a percolation parameter. We are interested in studying when there is a nontrivial percolation threshold in , namely, the critical percolation value . Our results for phase transition are analogous to those in the monopartite counterpart (see ). The similar (comparable) behavior of phase transitions in bipartite and corresponding monopartite networks has also been observed in other percolation contexts (see the discussion in Section 2).
Let be the size of a vertex set . The connected component containing the vertex is denoted by . By definition, the origin for all and . Since, for all and , and have the same distribution, it suffices to consider only without loss of generality. The percolation probability is defined as and the critical percolation value is defined as which is nondecreasing in for any given and .
Theorem 1. Assume that , and that . One has the following: (i)If , then .(ii)If , then .(iii)If , then . Moreover, there is almost surely at most one infinite component when .
Remark 2. The critical value turns out to be a function of only irrespective of the values of and . Koval et al.  showed the same behavior of for percolation in the monopartite lattice . This analogy of phase transition has been recognized in other percolation problems in statistical physics. An example is the percolation introduced by Mai and Halley  for the study of gelation processes. In this model, each vertex of an infinite connected graph is assigned one of two states, say and , with probability and , respectively, independently of all other vertices. Edges with two end-vertices having unlike states (called bonds) are occupied. Thus, the percolation can be viewed as a bond percolation with occupation probability (although some dependence is involved, namely, no odd path of bonds exists). Appel and Wierman  proved that percolation does not occur for any value of on a bipartite square lattice with bipartition and such that . In other words, the bond percolation cannot occur on the above bipartite square lattice for occupation probability . This is consistent with the classical result which says that bond percolation on does not occur when occupation probability (see, e.g., [15, 16]). Other comparable percolation thresholds for monopartite and bipartite high-dimensional lattices can be found in .
Another example is the biased percolation [18, 19] on infinite scale-free networks with a power-law degree distribution . In this model, an edge between vertices with degrees and is occupied with probability proportional to . By using generating function method, Hooyberghs et al.  showed that biased percolation on a bipartite scale-free network with two bipartition sets following degree distributions and , respectively, has the same critical behaviors with biased percolation on a monopartite scale-free network when .
Remark 3. The uniqueness of the infinite component holds here for the same reason as the uniqueness result for the percolation graph of (see [3, Theorem 2]). Note that our graph resulting from can be viewed as a spanning subgraph of that from .
We consider Theorem 1 as an intermediate step towards the study of percolation on bipartite hierarchical lattices. In particular, one may explore the connectivity at the critical regime and the graph distance (chemical distance) between and a vertex . It is also interesting to study the mean field percolation () and compare it with that on . Directed percolation  and other meaningful colorings on (other than the 2-coloring addressed in this paper) are possible.
We start with some notations. Then we prove Theorem 1.
For a vertex , define as the ball of radius around ; that is, . We make the following observations. Firstly, for any vertex , contains vertices. In particular, if , all vertices in the ball have the same type. Secondly, if . Finally, for any , , and , we have either or .
For a set of vertices, denote by its complement. Let be the component of vertices that are connected to by a path using only vertices within . For disjoint sets , we denote by the event that at least one edge joins a vertex in to a vertex in . means the event that such an edge does not exist. By definition, if for or , then occurs with probability 1. Let be the largest component in . If there are more than one such components, just take any one of them as . It is clear that .
Proof of (i). Let be the event that the origin connects by an edge to at least one vertex at distance in . By construction, for , . For , there are vertices in at distance from . Hence,
by using (5). For , there are vertices in at distance from . Similarly, we obtain
Since all the events are independent and diverges for any , , and , infinitely many of occur with probability 1 by the second Borel-Cantelli lemma. Thus, for any , , , and . The result then follows.
Proof of (ii). We only need to show by virtue of the monotonicity. Note that, for , there are vertices in and vertices in . Hence, by the comments in the proof of (i) and taking , for any , we obtain
which is strictly less than 1 for any finite .
Let and . We have Since the events are independent and all have the same probability strictly less than 1, Consequently, there exists an such that with probability 1. It follows from (12) that for all . This implies .
Proof of (iii). The positivity of is a direct consequence of the proof of Theorem 1(b) in . Since the percolation graph of can be viewed as a spanning subgraph of that of , the percolation cluster is almost surely finite; namely, , for small enough.
Now we turn to the proof of finiteness of . The main technique to be used is an iteration involving the tail probability of binomial distributions [3, 21]. Since , we choose an integer and a real number such that Clearly, . For , let and analogously, Here, is the probability that the largest component of a ball of radius contains at least vertices in and at least vertices in . Such a ball is said to be good. We set by convention. It is clear that, for , all and are positive, since is a finite number and the connection probability in (5) is positive.
In what follows, we will prove in two steps.
Step 1. We show that there exists some such that converges to 1 exponentially fast; namely, for some .
Step 2. We show that there exists some such that .
We start with Step 1. To this end, denote by the nonnegative integers. We can naturally label the vertices in via the map as This order agrees with the depiction in Figure 1. A ball of radius is said to be very good if it is good and its largest component connects by an edge to the largest component of the first (as per the aforementioned order) good subball in the same ball of radius . Clearly, the first good subball of radius in a ball of radius is very good. Condition (14) implies that . Thus we assert that the ball is good if (a) it contains good subballs of radius , and (b) all these good subballs are very good.
The number of good subballs of radius in a ball of radius has a binomial distribution with parameters and . Given the collection of good subballs, the probability that the first such good subball is very good equals to 1. Fix any of the other good subballs, say ; the probability that is not very good is upper bounded by where and are the number of vertices in the largest component of the first good subball in and , respectively; likewise, and are the number of vertices in the largest component of the subball in and , respectively. By definition, we have , , and the distance between two vertices in a ball of radius is at most .
Therefore, the probability for any of the other good subballs to be very good is at leat . Thus, the number of very good subballs is stochastically larger than a random variable obeying a binomial distribution . From the above comments (a) and (b) and the definition of , it follows that In general, we have the following inequality for the tail of binomial random variable: By (19), (20), and writing , we obtain We can choose large enough so that , and then we choose large enough so that (c) and (d) hold. To see (c), note that and then To see (d), note that , and by (21) we obtain which also approaches 0.
According to our above choice of and , we have inductively, if , then which implies that for all . We then finish the proof of Step 1.
For Step 2, recalling the definition of , we claim that In deed, if and , then is the first good subball in the derivation above. If this component is connected to at least other large components in as above, then the component containing the origin in has vertices in and vertices in . Thus, the inequality (25) follows.
A simple coupling gives Hence, we derive that the right-hand side of (26) converges to 1 exponentially fast by exploiting (20) and the fact that converges to 1 exponentially fast. It then follows from (25) that for some . Hence, It is direct to check that and hence for all . Since , inequality (28) yields as desired.
- D. A. Dawson and L. G. Gorostiza, “Percolation in a hierarchical random graph,” Communications on Stochastic Analysis, vol. 1, pp. 29–47, 2007.
- D. A. Dawson and L. G. Gorostiza, “Percolation in an ultrametric space,” Electronic Journal of Probability, vol. 18, article 12,, 2013.
- V. Koval, R. Meester, and P. Trapman, “Long-range percolation on the hierarchical lattice,” Electronic Journal of Probability, vol. 17, article 57, 2012.
- Y. Shang, “Percolation in a hierarchical lattice,” Zeitschrift Für Naturforschung A, vol. 67, pp. 225–229, 2012.
- S. R. Athreya and J. M. Swart, “Survival of contact processes on the hierarchical group,” Probability Theory and Related Fields, vol. 147, no. 3, pp. 529–563, 2010.
- Y. Shang, “A note on the perturbation of mixed percolation on the hierarchical group,” Zeitschrift Für Naturforschung A, vol. 68, pp. 475–478, 2013.
- T. Bojdecki, L. G. Gorostiza, and A. Talarczyk, “Number variance for hierarchical random walks and related fluctuations,” Electronic Journal of Probability, vol. 16, pp. 2059–2079, 2011.
- Y. Shang, “Mean commute time for random walks on hierarchical scale-free networks,” Internet Mathematics, vol. 8, pp. 321–337, 2012.
- H. Hooyberghs, B. Van Schaeybroeck, and J. O. Indekeu, “Percolation on bipartite scale-free networks,” Physica A, 2010.
- Y. Shang, “Estimation of the shortest average distance in bipartite networks with given density,” Journal of the Physical Society of Japan, vol. 80, no. 5, Article ID 055001, 2011.
- F. Liljeros, C. R. Edling, L. A. Nunes Amaral, H. E. Stanley, and Y. Åberg, “Social networks: the web of human sexual contacts,” Nature, vol. 411, no. 6840, pp. 907–908, 2001.
- M. E. J. Newman, “The structure of scientific collaboration networks,” Proceedings of the National Academy of Sciences, vol. 98, pp. 404–409, 2001.
- T. Mai and J. W. Halley, “AB percolation on a triangular lattice,” in Ordering in Two Dimensions, pp. 269–371, North-Holland, Amsterdam, 1980.
- M. J. Appel and J. C. Wierman, “On the absence of infinite AB percolation clusters in bipartite graphs,” Journal of Physics A, vol. 20, no. 9, article 36, pp. 2527–2531, 1987.
- B. Bollobás and O. Riordan, Percolation, Cambridge University Press, 2006.
- G. Grimmett, Percolation, Springer, New York, NY, USA, 1999.
- H. Kesten and Z. Su, “Some remarks on AB-percolation in high dimensions,” Journal of Mathematical Physics, vol. 41, no. 3, pp. 1298–1320, 2000.
- A. A. Moreira, J. S. Andrade Jr., H. J. Herrmann, and J. O. Indekeu, “How to make a fragile network robust and vice versa,” Physical Review Letters, vol. 102, no. 1, Article ID 018701, 2009.
- Y. Shang, “Biased edge failure in scale-free networks based on natural connectivity,” Indian Journal of Physics, vol. 86, pp. 485–488, 2012.
- Y. Shang, “Multi-type directed scale-free percolation,” Communications in Theoretical Physics, vol. 57, no. 4, pp. 701–716, 2012.
- F. M. Dekking and R. W. J. Meester, “On the structure of Mandelbrot's percolation process and other random cantor sets,” Journal of Statistical Physics, vol. 58, no. 5-6, pp. 1109–1126, 1990.
Copyright © 2013 Yilun Shang. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.