Vincent Corruble

An Emotional Model for Synthetic Characters with Personality

In recent years, emotional computing has found an important application domain in the field of interactive synthetic characters. Interesting examples of this domain are computer games, interface agents, human-robot interaction, etc. However, few systems in this area include a model of personality, although it plays an important role in differentiating agents and determining the way they experience emotions and the way they behave.

Discovering Rich Navigation Patterns on a Web Site

In this paper, we describe a method for discovering knowledge about users on a web site from data composed of demographic descriptions and site navigations. The goal is to obtain knowledge that is useful to answer two types of questions: (1) how do site users visit a web site? (2) Who are these users? Our approach is based on the following idea: the set of all site users can be divided into several coherent subgroups; each subgroup shows both distinct personal characteristics, and a distinct browsing behaviour. We aim at obtaining associations between site usage patterns and personal user descriptions. We call this combined knowledge ’rich navigation patterns’. This knowledge characterizes a precise web site usage and can be used in several applications: prediction of site navigation, recommendations or improvement in site design.

Dialogs taking into account experience, emotions and personality

This paper describes the DEEP project (Dialogs taking into account Experience, Emotions and Personality, adapted to computer games), which started in June 2006. The aim of the project is to provide generic solutions for the integration of autonomous Non Player Characters (NPCs) in next-generation adventure games. DEEP NPCs, equipped with a personality and a believable emotional engine, will use context-based information from the game environment and the player behavior to provide entertaining, rich and relevant dialogs.

A cognitive module in a decision-making architecture for agents in urban simulations

This paper addresses the issue of hybridization between reactive and cognitive approaches within a single decision-making architecture for virtual agent in an urban simulation. We use a reactive module in order to manage reactive behaviors and agent autonomy, and a cognitive module for anticipation, learning and complex behaviors management. The purpose of the cognitive module is to increase the agents behavior credibility. The agents reactive and proactive behaviors are sent to a decision module which is able to integrate, decompose, combine and select an action.

Helping Developers to Look Deeper inside Game Sessions

Game design and development activities are increasingly relying on the analysis of gamers behavior and preferences data. Various tools are available to the developers to track and analyze general data concerning acquisition, retention and monetization aspects of game commercialization. This is good enough to give hints on where problems are, but not to enable a precise diagnosis, which demands fine-grained data. For this kind of data, there is not enough support or guidance to decide which data to capture, to write the code to capture it, to choose the best representation of it and to allow an adequate retrieval and presentation of it. This paper introduces GameGuts (GG), a framework devoted to give further assistance to developers in choosing, representing, accessing and presenting game sessions fine-grained data. As a case study, GG recorded sessions of a game platform with over a hundred thousand users. The logs were analyzed using a Visual Domain Specific Language (as a query language) and an ensemble of rules (as a compliance test). The results are encouraging, since we could - among other results - find bugs and catch cheaters, as well as spot design flaws.

Extending the Strada Framework to Design an AI for ORTS

Strategy games constitute a significant challenge for game AI, as they involve a large number of states, agents and actions. This makes indeed the decision and learning algorithms difficult to design and implement. Many commercial strategy games use scripts in order to simulate intelligence, combined with knowledge which is in principle not accessible to human players, such as the position of the enemy base or the offensive power of its army. Nevertheless, recent research on adaptive techniques has shown promising results. The goal of this paper is to present the extension such a research methodology, named Strada , so that it is made applicable to the real-time strategy platform ORTS. The adaptations necessary to make Strada applicable to ORTS are detailed and involve the use of dynamic tactical points and specific training scenario for the learning AI. Two sets of experiments are conducted to evaluate the performances of the new method.

Simulating affective behaviours: an approach based on the COR theory

The expression of emotion is usually considered an important step towards the believability of a virtual agent. However, current models based on emotion categories face important challenges in their attempts to model the influence of emotions on agents behaviour. To adress this problem, we propose an architecture based on the COnservation of Resources theory (COR) which aims at producing affective behaviours in various scenarios. In this paper we explain the principle of such a model, how it is implemented and can be evaluated.

Evaluating the Impact of Anticipation on the Efficiency and Believability of Virtual Agents

We propose a model of cognitive process allowing virtual agents to exhibit anticipatory abilities. With user experiments, we show that this mechanism brings about an improvement in the efficiency of the behavior generated, and check that external observers are able to perceive it. We also confirm that this improvement in efficiency leads, up to a point, to an improvement in believability as judged by human observers. Beyond this level of efficiency, believability reaches a plateau.

Towards an Adaptive Supervision of Distributed Systems

The traditional, centralized, approach to supervision is challenged when communications between supervision and supervised systems become either slow, disrupted or too costly. To obtain a supervision system that is able to dynamically adapt itself to the communications’ state, we propose to distribute the supervision process through several autonomous agents. To evaluate our approach, we made experiments on a simulator for distributed systems using three different supervision approaches. Results show that our agent’s decision model does lead to a relevant autonomous supervision in distributed systems where a short response time prevails over a limited repair extra-cost.

An architecture for affective behaviour based on the conservation of resources

The display of emotions in virtual agents behaviours is usually recognized as an important element to enhance their believability. Therefore, applications aimed at realistic simulations or entertainment have to consider this aspect. This paper presents a model for autonomous virtual agents that enables them to adopt behaviours that can be perceived by human observers as emotional. Our goal is to obtain believable behaviours for several simulation scenari like a waiting line where conflicts can emerge between agents. The proposed architecture is based on a principle of conservation and acquisition of resources.