Artificial Intelligence

The striking progress of statistical learning over the past decade has led to wild expectations about the ability of machines to reproduce human behavior.

Labo bcom - intelligence artificielle
© Fred Pieau
These Artificial Intelligence techniques can apply equally to masses of data compiled in a computing center, or to data captured and processed by a smartphone. They are now entering a domestication phase, where technological, legal, and ethical questions are being asked in order to offer responsible AI approaches.

The Artificial Intelligence lab designs solutions to address issues in future networks, computer vision, cyberdefense, and e-health. Designing embedded imaging and radio communication applications, anomaly detection or classification, system resilience, human factor management, pathology prediction, and reliable diagnostics are the main challenges being tackled. To do so, the lab relies on traditional disciplines of automated language, signal, image processing, 3D vision, and dynamic programming, whose potential has been compounded by statistical learning. It can also count on its skills in hardware/software engineering and a legal team. Its methodological approach makes its algorithms explainable and robust so as to facilitate their certification.

Stéphane Paquelet - bcom
© Fred Pieau

Stéphane Paquelet

Artificial Intelligence lab Manager

With current learning techniques, we can automate tasks as varied as interpreting text, mapping space, generating realistic images, predicting behaviors, and planning actions.
products & services
Boost AI - Teaser produit - bcom b<>com *AI Coaching* services

Developing my competitiveness thanks to artificial intelligence

scientific publications


MES-loss: Mutually Equidistant Separation Metric Learning Loss Function

Deep metric learning has attracted much attention in recent years due to its extensive applications, such as clustering and image retrieval. Thanks to the success of deep learning (DL), many deep metric learning (DML) methods have been proposed. Neural networks (NNs) utilize DML loss functions to learn a mapping function that maps samples into a highly discriminative low-dimensional feature space…

read the publication


Channel charting based beamforming

Channel charting (CC) is an unsupervised learning method allowing to locate users relative to each other without reference. From a broader perspective, it can be viewed as a way to discover a low-dimensional latent space charting the channel manifold. In this paper, this latent modeling vision is leveraged together with a recently proposed location-based beamforming (LBB) method to show that…

read the publication


mpNet: variable depth unfolded neural network for massive MIMO channel estimation

Massive multiple-input multiple-output (MIMO) communication systems have a huge potential both in terms of data rate and energy efficiency, although channel estimation becomes challenging for a large number of antennas. Using a physical model allows to ease the problem by injecting a priori information based on the physics of propagation. However, such a model rests on simplifying assumptions and…

read the publication



Massive MIMO systems are highly efficient but critically rely on accurate channel state information (CSI) at the base station in order to determine appropriate precoders. CSI acquisition requires sending pilot symbols which induce an important overhead. In this paper, a method whose objective is to determine an appropriate precoder from the knowledge of the user's location only is proposed. Such…

read the publication