RFIA 2010

RFIA 2010

Conférenciers invités à la conférence RFIA 2010

Faculty of Science, University of Amsterdam

Human-centered Computing: Challenges and Perspectives

Human Centered Computing (HCC) is an emerging field that aims at bridging the existing gaps between the various disciplines involved with the design and implementation of computing systems that support people's activities. HCC aims at tightly integrating human sciences (e.g. social and cognitive) and computer science (e.g. human-computer interaction (HCI), signal processing, machine learning, and computer vision) for the design of computing systems with a human focus from beginning to end. This focus should consider the personal, social, and cultural contexts in which such systems are deployed.
In this presentation, I discuss the existing challenges in HCC and describe what I consider to be the three main areas of interest: media production, analysis (especially retrieval issues), and interaction. I will present my current research and how this is reflected into the HCC paradigm. In addition, I will identify the core characteristics of HCC, describe example applications, and propose a research agenda for HCC.

Schlomo Zilberstein

Department of Computer Science at the University of Massachusetts, Amherst.
Head of the The Resource-Bounded Reasoning Lab

Decentralized Decision Making: Challenges and New Directions

Coordinating the operation of a group of decision makers in stochastic environments is a long-standing challenge in AI. Decision theory offers a normative framework for optimizing decisions under uncertainty. But due to computational complexity barriers, developing decision-theoretic reasoning algorithms for multi-agent systems is a serious challenge. We describe a range of new formal models and algorithms to tackle this problem. Exact algorithms shed light on the structure and complexity of the problem, but they have limited use because only tiny problems can be solved optimally. We describe a number of effective approximation techniques that use bounded memory, sampling, and randomization. These methods can produce high-quality results in a variety of application domains such as mobile robot coordination and sensor network management. We examine the performance of these algorithms and describe current research efforts to further improve their applicability and scalability.

Olivier Teytaud

INRIA Saclay Ile-de-France, équipe projet TAO.

Prise de décision séquentielle dans l'incertain: le jeu de Go et au delà.

Le jeu de Go est devenu un challenge classique en intelligence artificielle, par sa grande dimension et sa complexité. En particulier, les humains restent considérablement meilleur que les ordinateurs. Néanmoins, l'écart s'est considérablement réduit ces dernières années grâce à des techniques nouvelles et généralistes, i.e. utilisables de manière générique pour des problèmes de décision séquentielle dans l'incertain.

Nous présenterons l'algorithme, appelé "Monte-Carlo Tree Search" (ou "Upper Confidence Trees" pour l'une de ses variantes), ses différences et ses ressemblances avec les algorithmes usuels, et ses forces et faiblesses, très apparentes sur le jeu de Go, ainsi que diverses applications loin des jeux.

Nous présenterons enfin un survol des formules dites de "bandit", abondamment utilisé comme brique de base dans le Monte-Carlo Tree Search.