Inference in Hidden Markov Models with Explicit State Duration Distributions
This work addresses the challenge of inference in EDHMMs for researchers and practitioners in fields like signal processing or bioinformatics, but it appears incremental as it adapts existing techniques to a specific model variant.
The authors tackled the problem of performing inference in Explicit-state-duration hidden Markov models (EDHMMs), which allow direct parameterization of state durations, by developing a tuning-parameter free, black-box inference procedure based on techniques from nonparametric HMMs, resulting in a method that avoids truncation or approximations typically required for such models.
In this letter we borrow from the inference techniques developed for unbounded state-cardinality (nonparametric) variants of the HMM and use them to develop a tuning-parameter free, black-box inference procedure for Explicit-state-duration hidden Markov models (EDHMM). EDHMMs are HMMs that have latent states consisting of both discrete state-indicator and discrete state-duration random variables. In contrast to the implicit geometric state duration distribution possessed by the standard HMM, EDHMMs allow the direct parameterisation and estimation of per-state duration distributions. As most duration distributions are defined over the positive integers, truncation or other approximations are usually required to perform EDHMM inference.