A Spectral Approach for Probabilistic Grammatical Inference on Trees

Abstract : We focus on the estimation of a probability distribution over a set of trees. We consider here the class of distributions computed by weighted automata - a strict generalization of probabilistic tree automata. This class of distributions (called rational distributions, or rational stochastic tree languages - RSTL) has an algebraic characterization: All the residuals (conditional) of such distributions lie in a finite-dimensional vector subspace. We propose a methodology based on Principal Components Analysis to identify this vector subspace. We provide an algorithm that computes an estimate of the target residuals vector subspace and builds a model which computes an estimate of the target distribution.
Document type :
Conference papers
Complete list of metadatas

Cited literature [18 references]  Display  Hide  Download

Contributor : Amaury Habrard <>
Submitted on : Thursday, July 7, 2011 - 7:48:49 PM
Last modification on : Friday, March 9, 2018 - 11:25:19 AM
Long-term archiving on : Monday, November 12, 2012 - 10:25:47 AM


Files produced by the author(s)


  • HAL Id : hal-00607096, version 1



Raphaël Bailly, Amaury Habrard, Francois Denis. A Spectral Approach for Probabilistic Grammatical Inference on Trees. 21st International Conference on Algorithmic Learning Theory (ALT 2010), Oct 2010, Australia. pp.74-88. ⟨hal-00607096⟩



Record views


Files downloads