A Novel Training Algorithm for HMMs with Partial and Noisy Access to the States
This addresses the challenge of improving HMM training for applications with imperfect state labels, though it appears incremental as an extension of existing EM methods.
The paper tackles the problem of estimating Hidden Markov Model (HMM) parameters when partial and noisy state information is available as side information, resulting in up to 70% improvement in state recognition performance compared to baseline algorithms.
This paper proposes a new estimation algorithm for the parameters of an HMM as to best account for the observed data. In this model, in addition to the observation sequence, we have \emph{partial} and \emph{noisy} access to the hidden state sequence as side information. This access can be seen as "partial labeling" of the hidden states. Furthermore, we model possible mislabeling in the side information in a joint framework and derive the corresponding EM updates accordingly. In our simulations, we observe that using this side information, we considerably improve the state recognition performance, up to 70%, with respect to the "achievable margin" defined by the baseline algorithms. Moreover, our algorithm is shown to be robust to the training conditions.