Pattern Recognition and Neural Networks
This 1996 book is a reliable account of the statistical framework for pattern recognition and machine learning. With unparalleled coverage and a wealth of case-studies this book gives valuable insight into both the theory and the enormously diverse applications (which can be found in remote sensing, astrophysics, engineering and medicine, for example). So that readers can develop their skills and understanding, many of the real data sets used in the book are available from the author's website: www.stats.ox.ac.uk/~ripley/PRbook/. For the same reason, many examples are included to illustrate real problems in pattern recognition. Unifying principles are highlighted, and the author gives an overview of the state of the subject, making the book valuable to experienced researchers in statistics, machine learning/artificial intelligence and engineering. The clear writing style means that the book is also a superb introduction for non-specialists.
Lietotāju komentāri - Rakstīt atsauksmi
Ierastajās vietās neesam atraduši nevienu atsauksmi.
Statistical Decision Theory
Linear Discriminant Analysis
Feedforward Neural Networks
Citi izdevumi - Skatīt visu
algorithm allow analysis applied approach approximation assume average Bayes bound called Chapter choose chosen classifier clusters combination components compute conditional consider continuous convergence covariance cross-validation decision density depend dimensions directed discriminant discussed distance distribution effect error error rate estimate et al example expected Figure functions given gives graph hidden idea independent inputs known layer learning least likelihood linear logistic marginal matrix maximize maximum mean measure methods minimize neural networks node normal Note observations optimal output parameters pattern performance points possible posterior probabilities predictive prior probability problem procedure projection properties Proposition random recognition regression rule sample scale selection separation shows smoothing space split squares statistical step suggests Suppose training set tree true units usually values variables variance vector vertices weights zero
Visi Grāmatu rezultāti »
Probabilistic Networks and Expert Systems: Exact Computational Methods for ...
Robert G. Cowell,Philip Dawid,Steffen L. Lauritzen,David J. Spiegelhalter
Priekšskatījums nav pieejams - 2003