Adaptively Growing Hierarchical Mixtures of Experts

Part of Advances in Neural Information Processing Systems 9 (NIPS 1996)

Bibtex Metadata Paper


Jürgen Fritsch, Michael Finke, Alex Waibel


We propose a novel approach to automatically growing and pruning Hierarchical Mixtures of Experts. The constructive algorithm pro(cid:173) posed here enables large hierarchies consisting of several hundred experts to be trained effectively. We show that HME's trained by our automatic growing procedure yield better generalization per(cid:173) formance than traditional static and balanced hierarchies. Eval(cid:173) uation of the algorithm is performed (1) on vowel classification and (2) within a hybrid version of the JANUS r9] speech recog(cid:173) nition system using a subset of the Switchboard large-vocabulary speaker-independent continuous speech recognition database.