Machine Learning: A Probabilistic Perspective

Capa
MIT Press, 24 de ago. de 2012 - 1104 páginas
A comprehensive introduction to machine learning that uses probabilistic models and inference as a unifying approach.

Today's Web-enabled deluge of electronic data calls for automated methods of data analysis. Machine learning provides these, developing methods that can automatically detect patterns in data and then use the uncovered patterns to predict future data. This textbook offers a comprehensive and self-contained introduction to the field of machine learning, based on a unified, probabilistic approach.

The coverage combines breadth and depth, offering necessary background material on such topics as probability, optimization, and linear algebra as well as discussion of recent developments in the field, including conditional random fields, L1 regularization, and deep learning. The book is written in an informal, accessible style, complete with pseudo-code for the most important algorithms. All topics are copiously illustrated with color images and worked examples drawn from such application domains as biology, text processing, computer vision, and robotics. Rather than providing a cookbook of different heuristic methods, the book stresses a principled model-based approach, often using the language of graphical models to specify models in a concise and intuitive way. Almost all the models described have been implemented in a MATLAB software package—PMTK (probabilistic modeling toolkit)—that is freely available online. The book is suitable for upper-level undergraduates with an introductory-level college math background and beginning graduate students.

 

Conteúdo

1
9
1
16
1
27
Generative models for discrete data
67
5
84
Gaussian models
92
Notation
99
Linear discriminant analysis LDA
105
Kernels
481
4
487
Gaussian processes
517
Adaptive basis function models
545
Markov and hidden Markov models
591
State space models
633
Undirected graphical models Markov random fields
663
Exact inference for graphical models
709

Nearest shrunken centroids classifier
111
4
121
5
128
Bayesian statistics
151
Frequentist statistics
193
Linear regression
219
Logistic regression
247
Generalized linear models and the exponential family
283
Directed graphical models Bayes nets
309
Mixture models and the EM algorithm
339
Latent linear models
383
Sparse linear models
423
Elastic net ridge and lasso combined
457
7
465
Learning a sparse coding dictionary
471
Variational inference
733
More variational inference
769
EP as a variational inference problem
790
Experimental comparison of graphcuts and
806
Monte Carlo inference
817
Clustering
877
Graphical model structure learning
909
Latent variable models for discrete data
949
Deep learning
999
Notation
1013
Bibliography
1019
Indexes
1051
839
1063
500
1067
Direitos autorais

Outras edições - Ver todos

Termos e frases comuns

Sobre o autor (2012)

Kevin P. Murphy is a Senior Staff Research Scientist at Google Research.

Informações bibliográficas