Variational Learning in Mixed-State Dynamic Graphical Models
This work addresses the challenge of modeling complex time-series data like gestures for applications in human-computer interaction, but it appears incremental as it builds on existing graphical model frameworks.
The paper tackled the problem of modeling real-valued stochastic time-series with both discrete and continuous causes, such as human hand gestures, by introducing a mixed-state dynamic graphical model that combines a hidden Markov model with a linear dynamic system, and they demonstrated its application in classifying human hand gestures made with a computer mouse.
Many real-valued stochastic time-series are locally linear (Gassian), but globally non-linear. For example, the trajectory of a human hand gesture can be viewed as a linear dynamic system driven by a nonlinear dynamic system that represents muscle actions. We present a mixed-state dynamic graphical model in which a hidden Markov model drives a linear dynamic system. This combination allows us to model both the discrete and continuous causes of trajectories such as human gestures. The number of computations needed for exact inference is exponential in the sequence length, so we derive an approximate variational inference technique that can also be used to learn the parameters of the discrete and continuous models. We show how the mixed-state model and the variational technique can be used to classify human hand gestures made with a computer mouse.