Deep learning delivers an interesting and conceptually as well as algorithmically
state-of-the-art approach to Artificial Intelligence and Intelligent Systems, in
general. This paradigm has been applied to numerous areas including machine
translation, computer vision, and natural language processing. Deep learning,
regarded as a subset of machine learning, utilizes a hierarchy level of artificial
neural networks to carry out efficiently the process of machine learning.
This volume provides the reader with a comprehensive and up-to-date treatise in
the area of deep learning. There are two focal and closely associated aspects here,
namely concepts supported by the environment of deep learning and a plethora of
its architectures. Those two faculties are strongly intertwined as well as linked with
the application domain under discussion. The concepts of deep learning revolve
around the structural and automatic (to a significant degree) mechanisms knowledge
representation. A variety of multilayer architectures bring about the tangible and
functionally meaningful pieces of knowledge. Their structural development
becomes essential to successful practical solutions. The architectural developments
that arise here support their detailed learning and refinements.
The chapters, authored by active researchers in the field, bring a collection of
studies that reflect upon the current trends in design and analysis of deep learning
topologies and ensuing applied areas of successful realizations including language
modeling, graph representation, and forecasting.
We would like to take this opportunity and express