How to couple from the past using a read-once source of randomness

  title={How to couple from the past using a read-once source of randomness},
  author={David Bruce Wilson},
  journal={Random Struct. Algorithms},
  • D. Wilson
  • Published 9 October 1999
  • Mathematics
  • Random Struct. Algorithms
We give a new method for generating perfectly random samples from the stationary distribution of a Markov chain. The method is related to coupling from the past (CFTP), but only runs the Markov chain forwards in time, and never restarts it at previous times in the past. The method is also related to an idea known as PASTA (Poisson arrivals see time averages) in the operations research literature. Because the new algorithm can be run using a read-once stream of randomness, we call it read-once… 
A kind of dual form for coupling from the past algorithm, to sample from Markov chain steady-state probability
This work proposes a kind of dual form for CFTP called D-CFTP that, in many situations, reduces the Monte Carlo simulation time and does not need to store the history of the used random numbers from one iteration to another.
The randomness recycler: a new technique for perfect sampling
  • J. A. Fill, M. Huber
  • Mathematics
    Proceedings 41st Annual Symposium on Foundations of Computer Science
  • 2000
A novel protocol, the randomness recycler (RR), that overcomes the difficulty of classical Markov chain approaches and gives the first expected linear time algorithms for generating samples.
Exact Sampling for Bayesian Inference: Unbounded State Spaces
Three modiications to CFTP are presented that make use of Meng's 7] multistage coupler to address the problem of uniform ergodic-ity in Markov chain Monte Carlo processes.
Relating Time and Customer Averages for Queues Using ‘forward’ Coupling from the Past
We give a new method for simulating the time average steady-state distribution of a continuous-time queueing system, by extending a ‘read-once’ or ‘forward’ version of the coupling from the past
Markov Chains and Mixing Times
This book is an introduction to the modern approach to the theory of Markov chains. The main goal of this approach is to determine the rate of convergence of a Markov chain to the stationary
Exact simulation of diffusions and new inference methods for discrete time data : also included, the one-shot CFTP algorithms
The research began by studying the Perfect Simulation idea developed by Propp and Wilson, and tried to apply Perfect Simulation ideas in the context of diffusion processes, coming up with the algorithm, termed the “Exact Algorithm” (EA), based simply on rejection sampling.
One-Shot CFTP; Application to a Class of Truncated Gaussian Densities
A new method for perfect simulation of multivariate densities by simulating efficiently from high-dimensional truncated normal distributions using the Gibbs sampler is introduced.
Perfect simulation refers to the art of converting suitable Markov Chain Monte Carlo algorithms into algorithms which return exact draws from the target distribution, instead of long-time
Perfect and nearly perfect sampling of work-conserving queues
Algorithms for perfect and nearly perfect sampling from the stationary distribution of the waiting times in various Poisson arrival multi-class and multi-server queues with non-preemptive work-conserving service disciplines are explored.
Perfect simulation of processes with long memory: A “coupling into and from the past” algorithm
We describe a new algorithm for the perfect simulation of variable length Markov chains and random systems with perfect connections. This algorithm, which generalizes Propp and Wilson's simulation


Exact sampling with coupled Markov chains and applications to statistical mechanics
This work describes a simple variant of this method that determines on its own when to stop and that outputs samples in exact accordance with the desired distribution, and uses couplings which have also played a role in other sampling schemes.
Perfect simulation and backward coupling
Algorithms for perfect or exact simulation of random samples from the invariant measure of a Markov chain have received considerable recent attention following the introduction of the
Exact Sampling for Bayesian Inference: Unbounded State Spaces
Three modiications to CFTP are presented that make use of Meng's 7] multistage coupler to address the problem of uniform ergodic-ity in Markov chain Monte Carlo processes.
An interruptible algorithm for perfect sampling via Markov chains
A new algorithm is presented which again uses the same Markov chains to produce perfect samples from n, but is baaed on a different idea (namely, acceptance/rejection sampling); and eliminates user-impatience bias.
On Arrivals That See Time Averages
It is shown that LBA, without the Poisson assumption, is necessary and sufficient for ASTA in a stationary process framework, which covers known examples of non-Poisson ASTA, such as certain flows in open Jackson queueing networks, as well as the familiar Poisson case PASTA.
Coupling from the past: A user's guide
The Markov chain Monte Carlo method is modified so as to remove all bias in the output resulting from the biased choice of an initial state for the chain, and this method is called Coupling From The Past (CFTP).
Perfect simulation of some point processes for the impatient user
  • E. Thönnes
  • Computer Science
    Advances in Applied Probability
  • 1999
An algorithm is constructed which protects against bias caused by user impatience and which delivers samples not only of the mixture model but also of the attractive area-interaction and the continuum random-cluster process.
How to Get a Perfectly Random Sample from a Generic Markov Chain and Generate a Random Spanning Tree of a Directed Graph
Algorithms for generating a random sample from the state space of a Markov chain in accordance with the steady-state probability law of the chain are given, improving on earlier results and exploiting the duality between the two problems.
Efficient exact sampling from the Ising model using Swendsen-Wang
This work gives the first method for running Swendsen-Wang as a completely coupling chain, thereby giving a procedure for taking exact samples from the ferromagnetic Ising model using Swendeden-wang.
Exact sampling and approximate counting techniques
  • M. Huber
  • Computer Science, Mathematics
    STOC '98
  • 1998
This work presents two algorithms for uniformly sampling from the proper colorings of a graph using colors using the Potts model, and presents the first polynomial time exact sampling algorithm for this problem, which is the first to run in polynometric time.