Latent Dependency Forest Models
This work addresses a bottleneck in probabilistic modeling for machine learning by introducing a novel method to handle context-specific independence, though it appears incremental as it builds on existing probabilistic frameworks.
The paper tackles the problem of modeling context-specific independence in probabilistic models by proposing latent dependency forest models (LDFMs), which use a dynamically changing forest structure to capture dependencies between random variables, and shows that LDFMs are competitive with existing models in experiments.
Probabilistic modeling is one of the foundations of modern machine learning and artificial intelligence. In this paper, we propose a novel type of probabilistic models named latent dependency forest models (LDFMs). A LDFM models the dependencies between random variables with a forest structure that can change dynamically based on the variable values. It is therefore capable of modeling context-specific independence. We parameterize a LDFM using a first-order non-projective dependency grammar. Learning LDFMs from data can be formulated purely as a parameter learning problem, and hence the difficult problem of model structure learning is circumvented. Our experimental results show that LDFMs are competitive with existing probabilistic models.