|
Simplifying Mixture Models with the Hierarchical EM Algorithm
|
| We propose a hierarchical EM algorithm for simplifying a finite mixture model into a reduced mixture model with fewer mixture components. The reduced model is obtained by maximizing a variational lower bound of the expected log-likelihood of a set of virtual samples. We develop four applications for our mixture simplification algorithm: recursive Bayesian filtering using Gaussian mixture model posteriors, KDE mixture reduction, belief propagation without sampling, and clustering hidden Markov models. For recursive Bayesian filtering, we propose an efficient algorithm for approximating an arbitrary likelihood function as a sum of scaled Gaussian. Experiments on synthetic data, human location modeling, visual tracking, vehicle self-localization, and eye gaze analysis show that our algorithm can be widely used for probabilistic data analysis, and is more accurate than other mixture simplification methods. |
|
Date: Jun 5 (Tue), 2018
|
|
Time: 12:30pm to 2:00pm
|
| Speaker: |
Dr. Antoni B. CHAN
( Bio )
Associate Professor,
Department of Computer Science,
City University of Hong Kong
|
| Venue: |
Room 14-221, 14/F, Lau Ming Wai Academic Building, City University of Hong Kong |
|
|
|
