Efficient and Effective Learning of HMMs Based on Identification of Hidden States

Figure 7

Identification performance for a 10-state discrete transient-cyclic HMM. See caption of Figure 6 for more details.

(a) Log-likelihoods during iterative training, BW with unknown

(b) Log-likelihoods during iterative training, BW given correct

(c) Heatmap of HMM model parameters