Date/heure
Date(s) - 13 mars 2015

Catégories Pas de Catégories

$Title : Distributional learning of some classes of context-free and midly \ncontext sensitive grammars.\n\nAbstract : This talk is concerned with the algorithmic learning of context-free and mildly context-sensitive languages.\nRecently, approaches generically called “distributional learning” have \nmade great success in the learning of CFLs.\nDistributional learning does Not learn probability distribution.\nDistributional learning is interested in which combination of a string \nand a context form a sentence belonging to the concerned CFLs, where a \ncontext is a pair of strings that wraps a string in the middle to form a \ncomplete sentence.\nWe will review some of the highlights of the development of \ndistributional learning theory, which may include the duality of the \nroles played by strings and contexts, hierarchy of distributionally \nlearnable classes, and generalization of distributional learning \ntechniques to mildly context-sensitive languages, which are proper but \nmild extensions of CFLs.[$