On Sampling Colorings of Bipartite Graphs

R. Balasubramanian (1) , C.R. Subramanian (1)

1 IMSc - Institute of Mathematical Sciences [Chennai]

We study the problem of efficiently sampling k-colorings of bipartite graphs. We show that a class of markov chains cannot be used as efficient samplers. Precisely, we show that, for any k, 6 ≤ k ≤ n^\1/3-ε \, ε > 0 fixed, \emphalmost every bipartite graph on n+n vertices is such that the mixing time of any markov chain asymptotically uniform on its k-colorings is exponential in n/k^2 (if it is allowed to only change the colors of O(n/k) vertices in a single transition step). This kind of exponential time mixing is called \emphtorpid mixing. As a corollary, we show that there are (for every n) bipartite graphs on 2n vertices with Δ (G) = Ω (\ln n) such that for every k, 6 ≤ k ≤ Δ /(6 \ln Δ ), each member of a large class of chains mixes torpidly. While, for fixed k, such negative results are implied by the work of CDF, our results are more general in that they allow k to grow with n. We also show that these negative results hold true for H-colorings of bipartite graphs provided H contains a spanning complete bipartite subgraph. We also present explicit examples of colorings (k-colorings or H-colorings) which admit 1-cautious chains that are ergodic and are shown to have exponential mixing time. While, for fixed k or fixed H, such negative results are implied by the work of CDF, our results are more general in that they allow k or H to vary with n.

Mots clés: Graph colorings; Markov chains; Analysis of algorithms

Domaines: Mathématique discrète [cs.DM]

R. Balasubramanian, C.R. Subramanian. On Sampling Colorings of Bipartite Graphs. Discrete Mathematics and Theoretical Computer Science, 2006, Vol. 8, pp.17-30. ⟨10.46298/dmtcs.368⟩. ⟨hal-00961100⟩