This part of the book summarizes the state of modern deep learning as it is used to
solve practical applications.
Deep learning has a long history and many aspirations. Several approaches have
been proposed that have yet to entirely bear fruit. Several ambitious goals have yet to
be realized. These less-developed branches of deep learning appear in the final part of
the book.
This part focuses only on those parts that are essentially working technologies that
are already used heavily in industry.
Modern deep learning provides a very powerful framework for supervised learning.
By adding more layers and more units within a layer, a deep network can represent
functions of increasing complexity. Most tasks that consist of mapping an input vector
to an output vector and are easy for a person to do can be accomplished via deep learning
given a large enough model and a large enough dataset of labeled training examples.
This part of the book describes this core parametric function approximation tech-
nology that is behind nearly all modern practical applications of deep learning. This
part of the book includes details such as how to efficiently model specific kinds of inputs,
such as how to process image inputs with convolutional networks and how to process
sequence inputs with recurrent and recursive networks. Moreover, we provide guidance
for how to preprocess the data for various tasks and how to choose the values of the
various settings that govern the behavior of these algorithms.
This part of the book is the most important for a practitioner–someone who wants
to begin implementing and using deep learning algorithms to solve real-world problems
today.
103
