Robust Mixing
Abstract
In this paper, we develop a new "robust mixing" framework for reasoning about adversarially modified Markov Chains (AMMC). Let $\mathbb{P}$ be the transition matrix of an irreducible Markov Chain with stationary distribution $\pi$. An adversary announces a sequence of stochastic matrices $\{\mathbb{A}_t\}_{t > 0}$ satisfying $\pi\mathbb{A}_t = \pi$. An AMMC process involves an application of $\mathbb{P}$ followed by $\mathbb{A}_t$ at time $t$. The robust mixing time of an ergodic Markov Chain $\mathbb{P}$ is the supremum over all adversarial strategies of the mixing time of the corresponding AMMC process. Applications include estimating the mixing times for certain non-Markovian processes and for reversible liftings of Markov Chains.
Non-Markovian card shuffling processes: The random-to-cyclic transposition process is a non-Markovian card shuffling process, which at time $t$, exchanges the card at position $L_t := t {\pmod n}$ with a random card. Mossel, Peres and Sinclair (2004) showed a lower bound of $(0.0345+o(1))n\log n$ for the mixing time of the random-to-cyclic transposition process. They also considered a generalization of this process where the choice of $L_t$ is adversarial, and proved an upper bound of $C n\log n + O(n)$ (with $C \approx 4\times 10^5$) on the mixing time. We reduce the constant to $1$ by showing that the random-to-top transposition chain (a Markov Chain) has robust mixing time $\leq n\log n + O(n)$ when the adversarial strategies are limited to holomorphic strategies, i.e. those strategies which preserve the symmetry of the underlying Markov Chain. We also show a $O(n\log^2 n)$ bound on the robust mixing time of the lazy random-to-top transposition chain when the adversary is not limited to holomorphic strategies.
Reversible liftings: Chen, Lovasz and Pak showed that for a reversible ergodic Markov Chain $\mathbb{P}$, any reversible lifting $\mathbb{Q}$ of $\mathbb{P}$ must satisfy $\mathcal{T}(\mathbb{P}) \leq \mathcal{T}(\mathbb{Q})\log (1/\pi_*)$ where $\pi_*$ is the minimum stationary probability. Looking at a specific adversarial strategy allows us to show that $\mathcal{T}(\mathbb{Q}) \geq r(\mathbb{P})$ where $r(\mathbb{P})$ is the relaxation time of $\mathbb{P}$. This gives an alternate proof of the reversible lifting result and helps identify cases where reversible liftings cannot improve the mixing time by more than a constant factor.
Full Text: Download PDF | View PDF online (requires PDF plugin)
Pages: 262-299
Publication Date: March 26, 2007
DOI: 10.1214/EJP.v12-398
References
- Aldous, David; Diaconis, Persi. Shuffling cards and stopping times. Amer. Math. Monthly 93 (1986), no. 5, 333--348. MR0841111 (88a:60021)
- Lee, Tzong-Yow; Yau, Horng-Tzer. Logarithmic Sobolev inequality for some models of random walks. Ann. Probab. 26 (1998), no. 4, 1855--1873. MR1675008 (2001b:60090)
- Ivona Bezakova and Daniel Stefankovic. Convex combinations of markov chains and sampling linear orderings. In preparation.
- Fang Chen, Laszlo Lovasz, and Igor Pak. Lifting markov chains to speed up mixing. In Proceedings of the thirty-first annual ACM symposium on Theory of computing, pages 275--281, New York, NY, USA, 1999. ACM Press. MR1798046
- Deepak Dhar. The abelian sandpile and related models. PHYSICA A, 263:4, 1999.
- Diaconis, P.; Saloff-Coste, L. Logarithmic Sobolev inequalities for finite Markov chains. Ann. Appl. Probab. 6 (1996), no. 3, 695--750. MR1410112 (97k:60176)
- Diaconis, Persi. Group representations in probability and statistics. Institute of Mathematical Statistics Lecture Notes---Monograph Series, 11. Institute of Mathematical Statistics, Hayward, CA, 1988. vi+198 pp. ISBN: 0-940600-14-5 MR0964069 (90a:60001)
- Diaconis, Persi; Saloff-Coste, Laurent. Comparison techniques for random walk on finite groups. Ann. Probab. 21 (1993), no. 4, 2131--2156. MR1245303 (95a:60009)
- Diaconis, Persi; Saloff-Coste, Laurent. Comparison theorems for reversible Markov chains. Ann. Appl. Probab. 3 (1993), no. 3, 696--730. MR1233621 (94i:60074)
- Diaconis, Persi; Shahshahani, Mehrdad. Generating a random permutation with random transpositions. Z. Wahrsch. Verw. Gebiete 57 (1981), no. 2, 159--179. MR0626813 (82h:60024)
- Gao, Fuqing; Quastel, Jeremy. Exponential decay of entropy in the random transposition and Bernoulli-Laplace models. Ann. Appl. Probab. 13 (2003), no. 4, 1591--1600. MR2023890 (2005a:60155)
- Goel, Sharad. Analysis of top to bottom-$k$ shuffles. Ann. Appl. Probab. 16 (2006), no. 1, 30--55. MR2209335 (Review)
- Hoeffding, Wassily. Probability inequalities for sums of bounded random variables. J. Amer. Statist. Assoc. 58 1963 13--30. MR0144363 (26 #1908)
- Horn, Roger A.; Johnson, Charles R. Topics in matrix analysis. Cambridge University Press, Cambridge, 1991. viii+607 pp. ISBN: 0-521-30587-X MR1091716 (92e:15003)
- Miclo, Laurent. Remarques sur l'hypercontractivité et l'évolution de l'entropie pour des chaînes de Markov finies. (French) [Remarks on hypercontractivity and evolution of entropy in finite Markov chains] Séminaire de Probabilités, XXXI, 136--167, Lecture Notes in Math., 1655, Springer, Berlin, 1997. MR1478724 (99b:60104)
- Mironov, Ilya. (Not so) random shuffles of RC4. Advances in cryptology---CRYPTO 2002, 304--319, Lecture Notes in Comput. Sci., 2442, Springer, Berlin, 2002. MR2054828 (2004m:94068)
- Ravi Montenegro. Duality and evolving set bounds on mixing times.
- Ravi Montenegro and Prasad Tetali. Mathematical Aspects of Mixing Times in Markov Chains, volume 1 of Foundations and Trends in Theoretical Computer Science. NOW Publishers, Boston-Delft, 2006.
- Elchanan Mossel, Yuval Peres, and Alistair Sinclair. Shuffling by semi-random transpositions. In Proceedings of FOCS 2004, volume 00, pages 572--581, Los Alamitos, CA, USA, 2004. IEEE Computer Society.
- Saloff-Coste, Laurent. Random walks on finite groups. Probability on discrete structures, 263--346, Encyclopaedia Math. Sci., 110, Springer, Berlin, 2004. MR2023654 (2004k:60133)

This work is licensed under a Creative Commons Attribution 3.0 License.