Consistency of Maximum Likelihood and Pseudo-Likelihood Estimators for Gibbs Distributions

By Basilis Gidas. 1988

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Sergey Ovchinnikov
2. Yoshua Bengio
3. Charles Sutton
4. Charu C. Aggarwal
5. Ruslan Salakhutdinov
6. Mohit Bansal
7. Jianfeng Gao
8. Quoc V. Le

Citations in the corpus

• Learning generative models for protein fold families (2011, background): “Moreover, this approximation is known to yield a consistent estimate of the parameters under fairly general conditions if the generating distribution is in fact a pairwise MRF defined by a graph over X [19].”
• Relational Dependency Networks (2007, background): “Gidas (1986) gives an alternative proof and Comets (1992) establishes a more general proof that does not require a finite state space x or stationarity of the true distribution Pθ0 .”
• Assessing the utility of coevolution-based residue–residue contact predictions in a sequence- and structure-rich era (2013, background): “Estimating the parameters by maximizing the pseudolikelihood is a consistent procedure (19).”
• Learning in Markov Random Fields with Contrastive Free Energies (2005, background): “Moreover it was shown that asymptotically this estimator is consistent [ 3 ] (but less efficient in the statistical sense than the MLE).”
• Finding and Leveraging Structure in Learning Problems (2012, background): “Moreover, this approximation is known to yield a consistent estimate of the parameters under fairly general conditions if the generating distribution is in fact a pairwise MRF defined by a graph over X [82].”
• Consistency of Pseudolikelihood Estimation of Fully Visible Boltzmann Machines (2006, background): “(2.3) Consistency of the pseudolikelihood has been thoroughly investigated for Markov random fields (see, e.g., Gidas, 1988, and Mase, 1995).”