Concentration bounds for unigram language models

MaRDI QIDQ3093298zbMATH OpenFDO

Authors Evgeny Drukh, Yishay Mansour

Publication date 12 October 2011

Full work available at URL http://www.jmlr.org/papers/v6/drukh05a.html

zbMATH Keywords

Chernoff boundsGood-Turing estimatorsleave-one-out estimationlogarithmic loss

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Natural language processing (68T50)

Recommendations

• Learning Theory
• Always Good Turing: asymptotically optimal probability estimation
• Concentration inequalities in the infinite urn scheme for occupancy counts and the missing mass, with applications
• On the concentration of the missing mass
• Maximum Likelihood Set for Estimating a Probability Mass Function

Cited in

(4)

• Learning Theory
• The layer complexity of Arthur-Merlin-like communication
• Generalized Good-Turing Improves Missing Mass Estimation
• Confidence Intervals for Parameters of Unobserved Events

This page was built for publication: Concentration bounds for unigram language models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3093298)