On the Importance Sampling of Self-Avoiding Walks
نویسنده
چکیده
In a 1976 paper published in Science, Knuth presented an algorithm to sample (non-uniform) self-avoiding walks crossing a square of side k. From this sample, he constructed an estimator for the number of such walks. The quality of this estimator is directly related to the (relative) variance of a certain random variable Xk . From his experiments, Knuth suspected that this variance was extremely large (so that the estimator would not be very efficient). But how large? For the analogous Rosenbluth algorithm, which samples unconfined self-avoiding walks of length n, the variance of the corresponding estimator is believed to be exponential in n. A few years ago, Bassetti and Diaconis showed that, for a sampler à la Knuth that generates walks crossing a k × k square and consisting of North and East steps, the relative variance is only O( √ k). In this note we take one step further and show that, for walks consisting of North, South and East steps, the relative variance jumps to 2k(k+1)/(k + 1)2k . This is quasi-exponential in the average length of the walks, which is of order k2. We also obtain partial results for general self-avoiding walks crossing a square, suggesting that the relative variance could be exponential in k2 (which is again the average length of these walks). Knuth’s algorithm is a basic example of a widely used technique called sequential importance sampling. The present paper, following Bassetti and Diaconis’ paper, is one of very few examples where the variance of the estimator can be found.
منابع مشابه
Stochastic Enumeration and Splitting Methods for Counting Self-Avoiding Walks
We present a new method for counting self-avoiding walks (SAW’s), called stochastic enumeration (SE), a stochastic replica of the naive (computationally intractable) full enumeration method. Also presented is a new approach, called OSLA-SPLIT, a combination of classic splitting with importance sampling. The latter is based on the well-known method, onestep-look-ahead (OSLA) method. Polynomial c...
متن کاملMulticanonical simulation of the Domb-Joyce model and the Go model: new enumeration methods for self-avoiding walks
We develop statistical enumeration methods for self-avoiding walks using a powerful sampling technique called the multicanonical Monte Carlo method. Using these methods, we estimate the numbers of the two dimensional N-step self-avoiding walks up to N = 256 with statistical errors. The developed methods are based on statistical mechanical models of paths which include self-avoiding walks. The c...
متن کاملGeneralised Atmospheric Rosenbluth Methods (GARM)
We show that the classical Rosenbluth method for sampling self-avoiding walks [1, 2] can be extended to a general algorithm for sampling many families of objects, including self-avoiding polygons. The implementation relies on an elementary move which is a generalisation of kinetic growth; rather than only appending edges to the endpoint, edges may be inserted at any vertex providing the resulti...
متن کاملSelf-avoiding walks, neighbour-avoiding walks and trails on semiregular lattices
We study self-avoiding and neighbour-avoiding walks and lattice trails on two semiregular lattices, the (3.122) lattice and the (4.82) lattice. For the (3.122) lattice we find the exact connective constant for both self-avoiding walks, neighbour-avoiding walks and trails. For the (4.82) lattice we generate long series which permit the accurate estimation of the connective constant for self-avoi...
متن کاملHow Many Needles Are in a Haystack, or How to Solve #P-Complete Counting Problems Fast
We present two randomized entropy-based algorithms for approximating quite general #P-complete counting problems, like the number of Hamiltonian cycles in a graph, the permanent, the number of self-avoiding walks and the satisfiability problem. In our algorithms we first cast the underlying counting problem into an associate rare-event probability estimation, and then apply dynamic importance s...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
- Combinatorics, Probability & Computing
دوره 23 شماره
صفحات -
تاریخ انتشار 2014