Building Social and Emotional Agents
Ana Paiva & Rui Prada
(Instituto Superior Tecnico, Portugal)
Socially Intelligent Agents are agents that have the ability to live and interact in a society inhabited by other agents, showing “human-style social intelligence”. Relying on foundational work coming from multi-agent systems, and combining it with aspects of human-agent interaction, this course will give an overview of how to build socially intelligent agents.
The course will start by providing an overview of agent societies and exploring the more traditional approaches related with such challenge. We will look at aspects of emergence, communication and cooperation in such societies. Then, we will investigate issues related with the design of social agents, in particular for embodied life-like characters and social robots. Some concrete examples will be presented and analysed.
The afternoon seminars will provide a way for students to explore deeper the material covered during the morning lessons.
Course Outline
Day 1: Introduction to Agents
Agents and Environments
Agent Architectures: Reactive, Deliberative and Hybrid
Societies and Encounters
Communication
Required Readings
Michael Wooldridge and Nick Jennings: "Agent Theories, Architectures, and Languages: A Survey" in Intelligent Agents, 1995.
Optional Readings
Michael Wooldrige: "An Introduction to Multi-Agent Systems" (Chapters 2 to 5). John Wiley and Sons, 2002.
Day 2: Agents in Society
Games and Strategies
Reaching an Agreement: Auctions and Negotiation
Cooperation
Emergent Behaviour
Required Readings
Robert Axelrod: "The Evolution of Strategies in the Iterated Prisoner's Dilemma". 1987.
Optional Readings
Michael Wooldrige: "An Introduction to Multi-Agent Systems" (Chapters 6, 7 and 9). John Wiley and Sons, 2002.
Day 3: Life-like Agents
Believability
Emotion and Personality
Facial and Body Expression
Required Readings
Joseph Bates "The role of emotion in believable agents" in Communications of the ACM - Volume 37, pages 122-125. July 1994.
Optional Readings
Diane Chi, Monica Costa, Liwei Zhao and Norman Badler: "The EMOTE model for effort and shape" in Proceedings of the 27th annual conference on Computer graphics and interactive techniques, pages 173 - 182. New York. 2000.
Day 4: Interacting with Users
Build Human-Agent Relationships
Group Interactions
Evaluation of Socially Intelligent Agents
Required Readings
Ana Paiva, Joao Dias, Daniel Sobral, Ruth Aylett, Polly Sobreperez, Sarah Woods, Carsten Zoll and Lynne Hall: "Caring for Agents and Agents that Care: Building Empathic Relations with Synthetic Agents" in Proceedings of the Third International Joint Conference on Autonomous Agents and Multiagent Systems - Volume 1, pages 194 - 201. New York. 2004
Ana Paiva, Rui Prada and Ana Paiva: "Believable Groups of Synthetic Characters" in Proceedings of the Fourth International Joint Conference on Autonomous Agents and Multiagent Systems, pages 37-43. The Netherlands. 2005.
Optional Readings
Kerstin Dautenhahn, Alan H. Bond, Lola Canamero and Bruce Edmonds (Eds): "Socially Intelligent Agents: Creating Relationships with Computers and Robots". Kluwer Academic Publishers, 2002.
Day 5: Applications of Socially Intelligent Agents
Socially Intelligent Agents in Games, Training and Education
No Required Readings