Department of Mathematical Sciences
|DATE:||Thursday, March 7, 2019|
|TIME:||1:15pm – 2:15pm|
|SPEAKER:||Yinsong Chen, Binghamton University|
|TITLE:||Image Analysis and Random Field with Simulated Annealing|
Markov Random Field is a common set up for image analysis and natural language processing. In analysis, Gibbs field was introduced into bayesian framework. Gibbs sampler is applied to sample from posterior distribution and simulated annealing method is applied to find the mode of posterior. For annealing, we will show its convergence and theoretical cooling schedule. At last, metropolis algorithm and conditional random field will be discussed for generalization and comparision.