| Document Type
|
:
|
BL
|
| Record Number
|
:
|
855246
|
| Main Entry
|
:
|
Cohen, Shay
|
| Title & Author
|
:
|
Bayesian analysis in natural language processing /\ Shay Cohen.
|
| Edition Statement
|
:
|
Second edition.
|
| Publication Statement
|
:
|
San Rafael :: Morgan & Claypool Publishers,, 2019.
|
| Series Statement
|
:
|
Synthesis lectures on human language technologies,; #41
|
| Page. NO
|
:
|
1 online resource (345 pages)
|
| ISBN
|
:
|
168173527X
|
|
|
:
|
: 9781681735276
|
|
|
:
|
1681735261
|
|
|
:
|
1681735288
|
|
|
:
|
9781681735269
|
|
|
:
|
9781681735283
|
| Notes
|
:
|
Part of: Synthesis digital library of engineering and computer science.
|
| Bibliographies/Indexes
|
:
|
Includes bibliographical references and index.
|
| Contents
|
:
|
Intro; List of Figures; List of Figures; List of Figures; Preface (First Edition); Acknowledgments (First Edition); Preface (Second Edition); Preliminaries; Probability Measures; Random Variables; Continuous and Discrete Random Variables; Joint Distribution over Multiple Random Variables; Conditional Distributions; Bayes' Rule; Independent and Conditionally Independent Random Variables; Exchangeable Random Variables; Expectations of Random Variables; Models; Parametric vs. Nonparametric Models; Inference with Models; Generative Models; Independence Assumptions in Models
|
|
|
:
|
Convergence Diagnosis
|
|
|
:
|
Decision-Theoretic Point EstimationDiscussion and Summary; Empirical Bayes; Asymptotic Behavior of the Posterior; Summary; Exercises; Sampling Methods; MCMC Algorithms: Overview; NLP Model Structure for MCMC Inference; Partitioning the Latent Variables; Gibbs Sampling; Collapsed Gibbs Sampling; Operator View; Parallelizing the Gibbs Sampler; Summary; The Metropolis-Hastings Algorithm; Variants of Metropolis-Hastings; Slice Sampling; Auxiliary Variable Sampling; The Use of Slice Sampling and Auxiliary Variable Sampling in NLP; Simulated Annealing; Convergence of MCMC Algorithms
|
|
|
:
|
Directed Graphical ModelsLearning from Data Scenarios; Bayesian and Frequentist Philosophy (Tip of the Iceberg); Summary; Exercises; Introduction; Overview: Where Bayesian Statistics and NLP Meet; First Example: The Latent Dirichlet Allocation Model; The Dirichlet Distribution; Inference; Summary; Second Example: Bayesian Text Regression; Conclusion and Summary; Exercises; Priors; Conjugate Priors; Conjugate Priors and Normalization Constants; The Use of Conjugate Priors with Latent Variable Models; Mixture of Conjugate Priors; Renormalized Conjugate Distributions
|
|
|
:
|
Discussion: To Be or not to Be Conjugate?Summary; Priors Over Multinomial and Categorical Distributions; The Dirichlet Distribution Re-Visited; The Logistic Normal Distribution; Discussion; Summary; Non-Informative Priors; Uniform and Improper Priors; Jeffreys Prior; Discussion; Conjugacy and Exponential Models; Multiple Parameter Draws in Models; Structural Priors; Conclusion and Summary; Exercises; Bayesian Estimation; Learning with Latent Variables: Two Views; Bayesian Point Estimation; Maximum a Posteriori Estimation; Posterior Approximations Based on the MAP Solution
|
|
|
:
|
Markov Chain: Basic TheorySampling Algorithms Not in the MCMC Realm; Monte Carlo Integration; Discussion; Computability of Distribution vs. Sampling; Nested MCMC Sampling; Runtime of MCMC Samplers; Particle Filtering; Conclusion and Summary; Exercises; Variational Inference; Variational Bound on Marginal Log-Likelihood; Mean-Field Approximation; Mean-Field Variational Inference Algorithm; Dirichlet-Multinomial Variational Inference; Connection to the Expectation-Maximization Algorithm; Empirical Bayes with Variational Inference; Discussion; Initialization of the Inference Algorithms
|
| Abstract
|
:
|
Natural language processing (NLP) went through a profound transformation in the mid-1980s when it shifted to make heavy use of corpora and data-driven techniques to analyze language. Since then, the use of statistical techniques in NLP has evolved in several ways. One such example of evolution took place in the late 1990s or early 2000s, when full-fledged Bayesian machinery was introduced to NLP. This Bayesian approach to NLP has come to accommodate various shortcomings in the frequentist approach and to enrich it, especially in the unsupervised setting, where statistical learning is done without target prediction examples. In this book, we cover the methods and algorithms that are needed to fluently read Bayesian learning papers in NLP and to do research in the area. These methods and algorithms are partially borrowed from both machine learning and statistics and are partially developed "in-house" in NLP. We cover inference techniques such as Markov chain Monte Carlo sampling and variational inference, Bayesian estimation, and nonparametric modeling. In response to rapid changes in the field, this second edition of the book includes a new chapter on representation learning and neural networks in the Bayesian context. We also cover fundamental concepts in Bayesian statistics such as prior distributions, conjugacy, and generative modeling. Finally, we review some of the fundamental modeling techniques in NLP, such as grammar modeling, neural networks and representation learning, and their use with Bayesian analysis.
|
| Subject
|
:
|
Bayesian statistical decision theory.
|
| Subject
|
:
|
Natural language processing (Computer science)
|
| Subject
|
:
|
Bayesian statistical decision theory.
|
| Subject
|
:
|
Natural language processing (Computer science)
|
| Dewey Classification
|
:
|
006.35
|
| LC Classification
|
:
|
QA76.9.N38C643 2019eb
|