Point of contact
Alexis Aravanis, L2S-CentraleSupelec
In the last few years, machine learning methods have been successfully used to realize tasks, e.g., image classification, speech recognition, translation of languages, that are usually simple to execute by human beings but are difficult to perform by machines. This is one of the reasons why machine learning is one of the main enablers to realize the notion of artificial intelligence. In order to identify the best machine learning solutions, e.g., the best architecture of an artificial neural network that allows one to fit input-output data pairs, the current methods consist of employing a data-driven approach. Once the artificial neural network is trained, it can respond to never-observed inputs by providing one with the optimum output based on past acquired knowledge. In this context, a recent trend in the machine community is to complement pure data-driven approaches with prior information based on expert knowledge. As opposed to other fields of science (e.g., image classification and speech recognition), in fact, mathematical models for communication networks optimization are very often available, even though they may be simplified and inaccurate. Therefore, this a priori expert knowledge, which has been acquired over decades of intense research, should not be dismissed and ignored. It is important, on the other hand, to develop new approaches that capitalize on the availability of (possibly simplified or inaccurate) theoretical models, in order to reduce the amount of empirical data to use and the complexity of training machine learning schemes.
Machine learning, deep neural networks, reinforcement learning.
- Alexis Aravanis, L2S-CentraleSupelec, email@example.com
- Marco Di Renzo, L2S-CentraleSupelec, firstname.lastname@example.org
- Merouane Debbah, Paris-Saclay University, email@example.com
- A. Zappone, M. Di Renzo, and M. Debbah, “Wireless Networks Design in the Era of Deep Learning: Model-Based, AI-Based, or Both?”, IEEE Transactions on Communications, Oct. 2019.
- A. Zappone, M. Di Renzo, M. Debbah et al., “Model-Aided Wireless Artificial Intelligence: Embedding Expert Knowledge in Deep Neural Networks Towards Wireless Systems Optimization,” IEEE Vehicular Technology Magazine, Sep. 2019.
- H. Gacanin and M. Di Renzo, “Wireless 2.0: Towards an Intelligent Radio Environment Empowered by Reconfigurable Meta-Surfaces and Artificial Intelligence,” IEEE Vehicular Technology Magazine, Dec. 2020.
- Emerging Technology Initiative on Machine Learning
- Best Readings in Machine Learning in Communications