IEEE International Conference on Communications
20-24 May 2019 // Shanghai, China
Empowering Intelligent Communications

W28: 2nd Workshop on Machine Learning in Wireless Communications (ML4COM)

W28: 2nd Workshop on Machine Learning in Wireless Communications (ML4COM)

Primary tabs

Machine learning techniques have been known for several decades, but only recent advances, both in terms of new methods and algorithms, and in terms of hardware capabilities, have allowed them to be widely and successfully implemented in various application areas. In particular, deep neural networks and their combination with reinforcement learning have allowed computers to reach superhuman levels in complex board games such as chess and go, or in popular Atari games, generating excitement among researchers of all disciplines.

The use of machine learning techniques in wireless comunications systems has already attracted a strong interest from our community with many proposed applications, both from academia and industry. In spite of some emerging applications,  it is not yet clear what will be the real gain of machine learning for wireless communication networks, and to which extent it will impact the design of future systems. Moreover, significant research efforts are needed  to transform these expectations into reality.

Another challenge lies in the fact that most of the design guidelines and results are only empirical, with a very scarce understanding of the theory behind the impressive observed performances. This is in contrast to the fundamental capacity results or error bounds we are often used in our community.  Improving the theoretical understanding of machine learning algorithms, and in particular of their applications to wireless communication problems is a key challenge that needs to be addressed in order to fully exploit the their potential. This challenge is particularly interesting for researchers in wireless communications as they have the theoretical background (e.g., from convex and distributed optimization to linear algebra, statistics, information theory, and coding) to contribute significantly to this research effort.

Investigating these research challenges and discussing the leading applications of machine learning in wireless communications constitute the core motivations of this workshop.