Composite Likelihood Estimation for Restricted Boltzmann machines
This work addresses parameter estimation difficulties in graphical models for researchers, but appears incremental as it builds on existing composite likelihood approximations.
The authors tackled the challenge of parameter learning in graphical models by proposing a composite likelihood method, which they applied to restricted Boltzmann machines, though no concrete numerical results were provided.
Learning the parameters of graphical models using the maximum likelihood estimation is generally hard which requires an approximation. Maximum composite likelihood estimations are statistical approximations of the maximum likelihood estimation which are higher-order generalizations of the maximum pseudo-likelihood estimation. In this paper, we propose a composite likelihood method and investigate its property. Furthermore, we apply our composite likelihood method to restricted Boltzmann machines.