Synopsis of `Computational Learning Theory: An Introduction'
This book concentrates on the `probably approximately correct' (PAC) model of learning, and gradually develop the ideas of efficiency considerations. Applications of the theory to artificial neural networks are briefly considered. Exercises are included throughout. The book is quite self-contained, as the necessary background material from logic, probability and complexity theory is included. It should therefore form an introduction to the theory of computational learning, suitable for a broad spectrum of graduate students from theoretical computer science and mathematics.
For a more detailed account of the PAC
model and its
generalizations, with an emphasis on applications to artificial neural
networks, see `Neural Network Learning: Theoretical
Foundations'.
Some reviews:
"notable for its clean, readable,
self-contained treatment of the foundations of PAC learnability theory"
Computing Reviews.
"The book is well-written: every notion
is
precisely described and sufficiently illustrated by examples." Zentralblatt MATH.
"The book is very well written. ... The
notions are illustrated with many examples and the proofs are presented
in a very clear way" Journal of Logic and Computation.