Recently\, deep learning has received significant attention as
a technique
to design and optimize wireless communication systems and n
etworks. The usual
approach to use deep learning consists of acquiring
large amount of empirical data
about the system behavior and employ it
for performance optimization. We believe\,
however\, that the applicati
on of deep learning to communication networks design
and optimization o
ffersmore possibilities. As opposed to other fields of science\, such
a
s image classification and speech recognition\, mathematical models for co
mmunication
networks optimization are very often available\, even thoug
h they may be
simplified and inaccurate.We believe that this a priori e
xpert knowledge\, which has
been acquired over decades of intense resea
rch\, cannot be dismissed and ignored.
In this tutorial\, in particular
\, we explore approaches that capitalize on the availability
of (possib
ly simplified or inaccurate) theoretical models\, in order to reduce
th
e amount of empirical data to use and the complexity of training artificia
l neural
networks (ANNs). We concretely show\, with the aid of some exa
mples\, that synergistically
combining prior expert knowledge based on
analytical models and datadriven
methods constitutes a suitable approac
h towards the design and optimization
of communication systems and netw
orks with the aid of deep learning based on
ANNs. The course is structu
red into three main parts:
• Deep learning by artificial neural netwo
rks: machine learning fundamentals\, feedforward
neural networks\, deep
reinforcement learning\, deep transfer learning
• Machine learning f
undamentals: feedforward neural networks\, deep reinforcement
learning\
, deep transfer learning
• Model-based design of wireless networks: p
erformance metrics\, convex optimization
techniques\, sequential optimi
zation techniques
• Model-aided deep learning for wireless communicat
ion design: learning to optimize\,
applications to interference network
s\, applications to massive MIMO networks\,
applications to energy-harv
esting in wireless networks
The tutorial will be mostly based o
n the following reference:
A. Zappone\, M. Di Renzo\, M. Debbah\, “Wi
reless Networks Design in the Era of Deep
Learning: Model-Based\, AI-Ba
sed\, or Both?”\, submitted to IEEE Transactions on
Communications\,
available at http://export.arxiv.org/abs/1902.02647
Bio: Alessio Zappone received his Master degree in Telecommunication E
ngineering
and his and Ph.D. degree in Electrical and Information Engin
eering from the
University of Cassino and Southern Lazio\, Cassino\, It
aly\, in 2007 and 2011\, respectively.
In 2009 he spent a semester at t
he Technische Universitaet Dresden\, Dresden\,
Germany\, as a visiting
Ph.D. student. In 2011/2012 he worked with Consorzio Nazionale
Interuni
versitario per le Telecomunicazioni (CNIT)\, in the framework of the
EU
FP7 project TREND. From 2012 to 2016 he has been the principal investigat
or
of the project CEMRIN\, carried out at the Chair of Communication Th
eory of the
Technische Universitaet Dresden\, and funded by the German
Research Foundation
(DFG). In 2014 he was the recipient of a Newcom mob
ility grant. In 2017 he has
been the recipient of the Marie Curie Indiv
idual Fellowship grant BESMART\, carried
out at the Large Networks and
Systems Group\, CentraleSupeléc\, Gif-sur-Yvette\,
France. His researc
h interests lie in the area of communication theory and signal
processi
ng\, with main focus on optimization techniques for resource allocation an
d
energy efficiency. Alessio serves as associate editor for the IEEE Si
gnal Processing
Letters and has served as associate editor for the IEEE
Journal on Selected Areas
on Communications (Special Issue on Energy-E
fficient Techniques for 5G Wireless
Communication Systems). He is the f
irst author of the monograph “Energy efficiency
in wireless networks
via fractional programming theory”\, appeared in the 2015
issue of Fo
undations and Trends in Communications and Information Theory. He
is an
IEEE Senior Member and a exemplary reviewer of IEEE Transactions on Commu
nications
and Transactions on Wireless Communications.