Dynamic Model Selection in IOHMMs

In this paper we describe adaptive model selection methods for an extension of the IOHMM called SimIOHMM. We show how to select the initial number of states of the HMM, how to decide when to add new states during the Baum-Welch iterations, and how to modify the Baum-Welch algorithm to efficiently add new nodes. We show that the SimIOHMM with dynamic model selection yields substantial computational gains over the IOHMM with no or little impact on predicting abilities.

By: Vittorio Castelli; Daniel A. Oblinger; Lawrence Bergman; Tessa A. Lau

Published in: RC23395 in 2004


This Research Report is available. This report has been submitted for publication outside of IBM and will probably be copyrighted if accepted for publication. It has been issued as a Research Report for early dissemination of its contents. In view of the transfer of copyright to the outside publisher, its distribution outside of IBM prior to publication should be limited to peer communications and specific requests. After outside publication, requests should be filled only by reprints or legally obtained copies of the article (e.g., payment of royalties). I have read and understand this notice and am a member of the scientific community outside or inside of IBM seeking a single copy only.


Questions about this service can be mailed to reports@us.ibm.com .