In this section, we present FTP results for graphs with bounded treewidth.
5.4.1. Maximal Cliques of a Graph with Bounded Treewidth
In this section, we present a systematic approach for generating all maximal cliques in a graph with bounded treewidth. Furthermore, we provide a comprehensive overview of the computation process for in each node i of the nice tree decomposition. These crucial steps serve as the foundation for developing our FPT algorithms for the d-FCTP, d-CTP, SCTP and MCTP using the dynamic programming technique.
Theorem 19. Assume that is a nice tree decomposition of a graph G with treewidth t. Let . The construction of for all can be accomplished in time.
Proof. By Theorem 18, there exists a bag
such that
for every
. The number of all maximal cliques of a graph with
n vertices is
and a list of all the maximal cliques can be generated in
time [
42]. Therefore, the construction of
can be achieved in
time for each node
.
Assume that T contains ℓ nodes. Following Lemma 16, we know that . Then, all can be constructed in time.
A maximal clique of is not necessarily a maximal clique of G. To construct , we start by an empty set . For each , we add it to if it is not a proper subset of any clique in for all nodes . Since , it takes time to check if one clique is a proper subset of another clique. Therefore, can be constructed in time. Then, all can be constructed in time. □
Lemma 17. Let be a nice tree decomposition of a graph G.
- (1)
Suppose that node i is a leaf node of T. Then, .
- (2)
Suppose that node i is a forget node of T. Let j be the child node of i and let such that . Then, .
- (3)
Suppose that node i is an introduce node of T. Let j be the child node of i and let such that . Let . Then, .
- (4)
Suppose that node i is a join node of T. Let j and ℓ be the child nodes of i Then, and .
Proof. - (1)
Since node i is a leaf node of G, we have . Therefore, .
- (2)
Since , we have . Therefore, .
- (3)
Since , we have . By Lemma 15, x is not a vertex of and thus for every . Hence, .
- (4)
Given that node i is a join node, we have . Consider two maximal cliques and . By Lemma 15, if , then is a subset of and is a subset of . We can assume that and .
If , it implies that and . In , there exists a maximal clique that properly contains and in , there also exists a maximal clique that properly contains . Therefore, and cannot be maximal cliques in G, which contradicts the definition of and . Thus, we conclude that and .
□
Theorem 20. Let be a nice tree decomposition of a graph G with treewidth t. Assume that T has nodes. The construction of all can be accomplished in time.
Proof. We start from the leaves in T up to the root, computing the solutions for each visited node i on the way through the dynamic programming technique. By Lemma 17, can be computed in constant time if node i is a leaf node or a forget node. We consider the following cases for introduce and join nodes.
Case 1: Node i is an introduce node. Let and . Lemma 17 shows that . Note that for every . To obtain Q, we have to check each clique of to see if it contains the vertex x. Therefore, can be computed in time.
Case 2: Node i is a join node. Let j and ℓ be the child nodes of i. Lemma 17 shows that . To eliminate all cliques of from , it is necessary to check each maximal clique to see if . Notably, a maximum clique contains at most vertices and determining set equality takes time. Therefore, the computation of can be achieved in time.
We mentioned earlier in the proof of Theorem 19 that for each node p, contains at most maximal cliques of G and contains at most maximal cliques of G. Additionally, by Lemma 16, the decomposition tree contain at most vertices. We have .
Considering the above analysis, the construction of all can be accomplished in time. □
In the following, we are going to introduce the -clique transversal problem. The problem includes the d-FCTP, -CTP, SCTP and MCTP as special cases. That gives us a unified approach to the four problem by solving the -clique transversal problem.
5.4.2. The -clique Transversal Problem
In this section, we introduce the -clique transversal problem, which serves as a unifying framework encompassing the d-FCTP, d-CTP, SCTP and MCTP as specific instances. By formulating these four problems within the context of the -clique transversal problem, we can adopt a unified approach to address all of them. This approach not only simplifies the problem-solving process but also allows us to leverage shared problem structures and solution techniques, leading to more efficient and effective solutions for each individual problem.
Definition 12. Suppose that G is a graph and and are fixed. Let . A function is a -clique transversal function ((Y,z)-CTF) of G if for every . The -clique transversal number of G, denoted by , is the minimum weight of a -CTF of G. The -clique transversal problem (-CTP) is to find a minimum -CTF of G.
Definition 13. Suppose that and are fixed. Let for and let . Let G be a graph and let z be an integer in Y. Let be a p-tuple of subsets of . The weight of the p-tuple X, denoted by , is . Let and let . Let be another p-tuple of subsets of . We give the following notations and definitions.
- 1.
denotes the p-tuple .
- 2.
denotes the p-tuple .
- 3.
denotes the p-tuple such that and for .
- 4.
denotes the p-tuple .
- 5.
denotes the p-tuple such that and for .
- 6.
A p-tuple is a p-partition of satisfying the following conditions.
- (a)
.
- (b)
for .
- 7.
A p-assignment of is a p-partition of such that for every
Remark 1. A -CTF of G can be regarded as an p-assignment of and vice versa. Then, is a p-assignment of .
Definition 14. Suppose that and are fixed. Let for and let . Let G be a graph of bounded treewidth with a nice tree decomposition rooted at node r and let .
For each node , let be a p-partition of and let be a p-assignment of . The p-assignment is called a node p-assignment if satisfies all the following conditions.
- (1)
For each , and .
- (2)
.
- (3)
for every .
Let be a node p-assignment of of minimum weight. If does not exist, let with .
Remark 2. By Definition 14, , and is a p-partition of .
Lemma 18. Let be a nice tree decomposition of a graph G with treewidth t. Suppose that node i is a leaf node of T. For all p-partitions X of , the node p-assignments of can be computed in time.
Proof. Let and . Given that node i is a leaf node, it follows that and . Consequently, we have and for .
In the scenario where is an empty set, we can conclude that for every p-partition X of . Thus, we proceed under the assumption that .
There are at most maximal cliques in . The number of p-partitions of is bounded by . For each , the verification process to determine whether can be performed in time by the summation of the weights over the vertices in C.
By definition of tree decomposition, we establish . Consequently, considering the preceding discussion, all node p-assignments of can be computed in time. □
Lemma 19. Let be a nice tree decomposition of a graph G with treewidth t. Suppose that node i is a forget node of T. Let j be the child node of i and let such that . Let X be a p-partition of and . Let such that . Assume that and , where . Let be a p-assignment of such that , and for every . Then, is a node p-assignment of and .
Proof. Let and . Since , and there exists an integer such that .
Let and let such that . Assume that and , where . We can obtain a node p-assignment of by , and for every . Then, .
Conversely, we assume that and . We can obtain a node p-assignment of by , and for every . Then, . Following the discussion above, the lemma holds. □
Lemma 20. Let be a nice tree decomposition of a graph G with treewidth t. Suppose that node i is an introduce node of T. Let j be the child node of i and let x be the vertex such that . Let . Let and let be a p-partition of such that . Then, Proof. Since , we have by Lemma 15. We consider the following two cases.
Case 1: . Then, . Let be a p-partition of . Let and let X be a p-partition of such that . Then, is a node p-assignment of such that . Therefore, .
Conversely, let be a p-partition of such that . Then, is a p-assignment of such that . Let . Therefore, . Based on the discussion above, we have .
Case 2: . Lemma 17 shows that . Let be a p-partition of . Let and let X be a p-partition of such that .
If or there exists a maximal clique such that , then . Otherwise, is a p-assignment S of such that . Therefore, .
Conversely, let be a p-partition of such that . If , then is a p-assignment of such that . Let . Therefore, . Based on the discussion above, we have . □
Lemma 21. Let G be a graph of bounded treewidth with a nice tree decomposition . Suppose that node i is a join node of T. Let j and ℓ be the child nodes of i. For each p-partition of , Proof. Since node i is a join node, . Then, . Let . Clearly, and for . We have .
Let and . Then, L and R are node p-assignments of and , respectively. Furthermore, . Therefore, .
In light of the preceding discussion, we conclude that , thereby establishing the validity of the lemma. □
The -CTP in an n-vertex graph G with a nice tree decomposition of treewidth t can be efficiently solved. We present the following theorem:
Theorem 21. Under the assumption that has been computed for each node i in T, the -CTP can be solved in time.
Proof. Assuming that the tree T is rooted at r, we have and . Our algorithm follows a bottom-up approach, starting from the leaves in T and correctly computing the solutions for each visited node i using dynamic programming techniques. This process allows us to determine , which is the minimum weight among all solutions for p-partitions X of .
According to Lemma 18, if a node i is a leaf node in T, we can compute the node p-assignments of for all p-partitions X of in time.
Similarly, if a node i is a forget node in T, by Lemma 19, we can compute the node p-assignments of for all p-partitions X of in time.
When dealing with an introduce node i in T, we need to compute and verify if for every . This computation and verification can be done in time. Consequently, the node p-assignments of for all p-partitions X of can be computed in time.
Regarding join nodes, as described in Lemma 21, if a node i is a join node in T and , we can assume that and . Therefore, . By considering the proof of Lemma 21, we have and for . Consequently, . The computation of can be achieved in time, allowing us to compute the node p-assignments of for all p-partitions X of in time.
Considering that T contains nodes, the -CTP can be solved in time. □
Corollary 4. Let be a nice tree decomposition of an n-vertex graph G with treewidth t. Under the assumption that has been computed for each node i in T, the following results hold:
- (1)
The d-FCTP can be solved in time.
- (2)
The SCTP can be solved in time.
- (3)
The MCTP can be solved in time.
- (4)
The -CTP can be solved in time.
Proof. These results can be derived by recognizing that the d-FCTP, SCTP, MCTP and d-CTP can all be viewed as special cases of the -CTP problem. By assigning specific values to for each problem, we can establish a connection between them.
For instance, consider the d-FCTP. It corresponds to the special case of the -CTP where and . Theorem 21 provides a general time complexity result for solving the -CTP, stating that it can be solved in time. In the case of the d-FCTP, we have . Therefore, the d-FCTP can be solved in time, which simplifies to and further to .
Similarly, we can apply this approach to the SCTP, MCTP and d-CTP by setting the corresponding values of as follows: for the SCTP, for the MCTP and for the -CTP. By mapping these problems to the more general -CTP framework, we can leverage the time complexity results established for the -CTP to determine the respective complexities of the SCTP, MCTP and d-CTP. □
Theorem 22 ([
1]).
An n-vertex planar graph with a domination number of k has a treewidth of at most . Additionally, a tree decomposition of such a graph can be found in time. Corollary 5. Let G be a planar graph with n vertices. The following results hold.
- (1)
If , it can be computed in time.
- (2)
If for or , it can be computed in time.
- (3)
If , the -CTP can be solved in time.
Proof. - (1)
Let and . We consider the following two cases.
Case 1: . In this case, G does not have any vertex with degree larger than . Therefore, and as shown by Lemma 9 and Theorem 11.
Since a CTS of G is also a dominating set of G, we have . By Theorem 22, we can construct a tree decomposition of G with the width at most in time.
Let
and let
be a nice tree decomposition of a graph
G with the width
t. Assume that
T has
ℓ nodes. By Theorem 20, we can construct
in time
Following the arguments in Corollary 4, we can compute
in time
If we add the time for construction of
to the computation of
, the total running time is:
Case 2: . Lemma 8 implies that we must include all vertices in when constructing a CTS of G of size at most k. Our objective is then to find a CTS D of size at most such that every maximal clique has a vertex in D. In other words, the set D is a CTS of the clique subgraph of size at most , where . Ultimately, the set forms a CTS of G of size at most k. By Theorem 11, the construction of and can be obtained in time. Following the discussion in Case 1, D can be found in time.
Hence, if , it can be computed in time.
- (2)
Assume that the maximum clique size of G is . We consider the d-FCTP on G for or . By Lemma 11, we have . Clearly, . By Theorem 22, we can construct a tree decomposition of G with the width at most in time. Following similar arguments to those in statement (1) with Lemma 11, we can conclude a running time of .
- (3)
Clearly, . By Theorem 22, we can construct a tree decomposition of G with the width at most in time. Following similar arguments to those in statement (1) with Lemmas 12 and 13, we conclude a running time of . This completes the proof of the corollary.
□