Learning Tree Distributions by Hidden Markov Models
This work addresses the challenge of interpretable distribution learning for tree-structured data, but it appears incremental as it builds on existing hidden tree Markov models with a novel generalization.
The paper tackles the problem of learning distributions for tree-structured data using hidden tree Markov models, focusing on causality assumptions from tree visit direction, and introduces a non-parametric generalization of the bottom-up model interpreted as an infinite-state nondeterministic tree automaton.
Hidden tree Markov models allow learning distributions for tree structured data while being interpretable as nondeterministic automata. We provide a concise summary of the main approaches in literature, focusing in particular on the causality assumptions introduced by the choice of a specific tree visit direction. We will then sketch a novel non-parametric generalization of the bottom-up hidden tree Markov model with its interpretation as a nondeterministic tree automaton with infinite states.