A maximum likelihood algorithm for the estimation and renormalization of exponential densities
arXiv:math/0409230 · doi:10.1016/j.jcp.2005.03.001
Abstract
We present an algorithm based on maximum likelihood for the estimation and renormalization (marginalization) of exponential densities. The moment-matching problem resulting from the maximization of the likelihood is solved as an optimization problem using the Levenberg-Marquardt algorithm. In the case of renormalization, the moments needed to set up the moment-matching problem are evaluated using Swendsen's renormalization method. We focus on the renormalization version of the algorithm, where we demonstrate its use by computing the critical temperature of the two-dimensional Ising model. Possible applications of the algorithm are discussed.
18 pages, 3 figures. Submitted to the Journal of Computational Physics
Cited by in corpus (7)
- Recent Advances in Maximum Entropy Biasing Techniques for Molecular Dynamics
- Efficient Monte Carlo sampling by parallel marginalization
- Molecular simulations minimally restrained by experimental data
- Problem reduction, renormalization, and memory
- Parallel marginalization Monte Carlo with applications to conditional path sampling
- Stochastic global optimization as a filtering problem
- Phase transition approach to detecting singularities of PDEs