Co-localisées avec la Plate-Forme Intelligence Artificielle (PFIA 2019)International audienceTo provide artificial intelligence with better adaptability, we suggest a coupling based on an enactivist and cognitivist approach between deep neural networks and learning classifier systems. As autonomy is a vector for the development of adaptability, a definition is first proposed to support later the choices and contributions of perception, reasoning and memory models used. Those models, namely these networks, these systems and a learning tree we design, that keeps track of the experiences of the coupled system, should allow these intelligences to identify, by themselves, problems arising from their environment and design in complet...
An intelligent agent immerged in its environment must be able to both understand andinteract with th...
By using learning mechanisms extracted from recent discoveries in neuroscience, spiking neural netwo...
International audienceProgress in machine learning, more recently accelerated by deep learning, now ...
The most powerful artificial intelligence systems are now based on learned statisticalmodels. In ord...
The brain, beyond its primary sensori-motor and regulation functions, is an outstanding adaptive sys...
If the perception of our environment seems easy and natural, the visual attributes of various object...
With the appearance of the paradigms of ubiquitous systems and ambient intelligence, a new domain of...
Nous modifions l’algorithme non supervisé de Kohonen sur la base de considérations biologiques, dans...
In the following, we study the contribution of information technology (IT) environments on human lea...
With the appearance of the paradigms of ubiquitous systems and ambient intelligence, a new domain of...
Robotics has reached a high accuracy on many tasks, like for instance manipulation or navigation. Bu...
This thesis aims to propose an architecture that addresses the design, supervision, management and a...
L'article traite de la modélisation cognitive de l'étudiant à la base de la conception des systèmes ...
International audienceL'intelligence artificielle (IA) a connu ces dernières années de grandes avanc...
Co-localisées avec la Plate-Forme Intelligence Artificielle (PFIA 2019)National audienceDesign of di...
An intelligent agent immerged in its environment must be able to both understand andinteract with th...
By using learning mechanisms extracted from recent discoveries in neuroscience, spiking neural netwo...
International audienceProgress in machine learning, more recently accelerated by deep learning, now ...
The most powerful artificial intelligence systems are now based on learned statisticalmodels. In ord...
The brain, beyond its primary sensori-motor and regulation functions, is an outstanding adaptive sys...
If the perception of our environment seems easy and natural, the visual attributes of various object...
With the appearance of the paradigms of ubiquitous systems and ambient intelligence, a new domain of...
Nous modifions l’algorithme non supervisé de Kohonen sur la base de considérations biologiques, dans...
In the following, we study the contribution of information technology (IT) environments on human lea...
With the appearance of the paradigms of ubiquitous systems and ambient intelligence, a new domain of...
Robotics has reached a high accuracy on many tasks, like for instance manipulation or navigation. Bu...
This thesis aims to propose an architecture that addresses the design, supervision, management and a...
L'article traite de la modélisation cognitive de l'étudiant à la base de la conception des systèmes ...
International audienceL'intelligence artificielle (IA) a connu ces dernières années de grandes avanc...
Co-localisées avec la Plate-Forme Intelligence Artificielle (PFIA 2019)National audienceDesign of di...
An intelligent agent immerged in its environment must be able to both understand andinteract with th...
By using learning mechanisms extracted from recent discoveries in neuroscience, spiking neural netwo...
International audienceProgress in machine learning, more recently accelerated by deep learning, now ...