Description du projet
Un partenariat harmonieux entre l’homme et l’IA
Dans le domaine de l’intelligence artificielle, il est primordial de relever le défi complexe de la méfiance des utilisateurs. Ce scepticisme est particulièrement marqué lorsqu’il s’agit de problèmes de prise de décision séquentielle. Ces applications d’IA impliquent des séquences d’actions complexes, où la transparence et les solutions sur mesure font souvent défaut, ce qui entraîne des interprétations erronées, des hésitations et un manque général d’acceptation de la part des utilisateurs. Dans ce contexte, le projet PEER, financé par l’UE, vise à remodeler le paysage de la collaboration entre l’homme et l’IA. L’approche pionnière de PEER consiste à placer les utilisateurs au cœur du processus d’IA, en favorisant la communication bidirectionnelle et l’apprentissage mutuel. Faisant le lien entre les sciences sociales et l’IA, PEER élaborera de nouvelles méthodes d’engagement, concevra des techniques de planification de l’IA pour des scénarios dynamiques, introduira un indice d’acceptation de l’IA et testera ces différents concepts.
Objectif
A significant, highly complex class of artificial intelligence applications are sequential decision-making problems, where a sequence of actions needs to be planned and taken to achieve a desired goal. Examples include routing problems, which involve a sequence of steps from source to destination; the control of manufacturing processes, which consist of a variable sequence of operations; or active learning problems, in which machine learning algorithms query human users for a sequence of inputs.
We address the compelling scientific and technological goal of tackling users' lack of trust in AI, which currently often hinders the acceptance of AI systems. We break down this problem into two complementary aspects. First, users do not understand current AI systems well, with a lack of transparency leading to misinterpretations and mistrust. Second, current AI systems do not understand users well, offering solutions that are inadequately tailored to the users' needs and preferences.
PEER will focus on how to systematically put the user at the centre of the entire AI design, development, deployment, and evaluation pipeline, allowing for truly mixed human-AI initiatives on complex sequential decision-making problems. The central idea is to enable a two-way communication flow with enhanced feedback loops between users and AI, leading to improved human-AI collaboration, mutual learning and reasoning, and thus increased user trust and acceptance. As an interdisciplinary project between social sciences and artificial intelligence, PEER will facilitate novel ways of engagement by end-users with AI in the design phase; will create novel AI planning methods for sequential settings which support bidirectional conversation and collaboration between users and AI; will develop an AI acceptance index for the evaluation of AI systems from a human-centric perspective; and will conduct an integration and evaluation of these novel approaches in several real-world use cases.
Champ scientifique
Programme(s)
Thème(s)
Régime de financement
HORIZON-RIA - HORIZON Research and Innovation ActionsCoordinateur
1050 Bruxelles / Brussel
Belgique