Description du projet
La modélisation informatique pour mieux comprendre l’utilisation du langage humain
Modéliser la capacité de communiquer en utilisant un langage grâce à des méthodes informatiques constitue un problème ouvert fascinant. Afin de relever ce défi, le projet DREAM, financé par l’UE, créera de nouveaux modèles informatiques d’agents de dialogue qui apprennent comment engager le dialogue directement à partir de données impliquant l’utilisation du langage. Le projet associe des idées issues de la linguistique avec de nouvelles avancées dans l’apprentissage automatique lié aux systèmes de réseaux de neurones artificiels. Cela permettra aux agents de dialogue d’apprendre les représentations qu’ils manipulent directement à partir de l’expérience. Ils apprendront à partir de conversations humaines ayant une finalité orientée sur les tâches. Le projet permettra de faire avancer le développement d’agents de conversation complexes et d’améliorer la compréhension scientifique de l’utilisation du langage humain.
Objectif
Our ability to communicate using language in conversation is considered the hallmark of human intelligence. Yet, while holding a dialogue is effortless for most of us, modelling this basic human skill by computational means has proven extremely difficult. In DREAM, I address this challenge by establishing a new computational model of a dialogue agent that can learn to take part in conversation directly from data about language use. DREAM stands at the crossroads of the symbolic and the sub-symbolic traditions regarding the nature of human cognitive processing and, by extension, its computational modelling. My model is grounded in linguistic theories of dialogue, rooted in the symbolic tradition, but exploits recent advances in computational learning that allow the agent to learn the representations that it manipulates, which are distributed and sub-symbolic, directly from experience. This is an original approach that constitutes a paradigm shift in dialogue modelling --- from predefined symbolic representations to automatic representation learning --- that will break new scientific ground in Computational Linguistics, Linguistics, and Artificial Intelligence. The DREAM agent will be implemented as an artificial neural network system and trained with task-oriented conversations where the participants have a well-defined end goal. The agent will be able to integrate linguistic and perceptual information and will be endowed with the capability to dynamically track both speaker commitments and partner-specific conventions, leading to more human-like and effective communication. Besides providing a breakthrough in our capacity to build sophisticated conversational agents, DREAM will have substantial impact on our scientific understanding of human language use, thanks to its emphasis on theory-driven hypotheses and model analysis.
Champ scientifique
Programme(s)
Régime de financement
ERC-COG - Consolidator GrantInstitution d’accueil
1012WX Amsterdam
Pays-Bas