Prof. Ana Paiva

I’m a Full Professor in the Department of Computer Engineering, IST (“Instituto Superior Técnico”) from the University of Lisbon and coordinator of GAIPS – “Intelligent and Social Agents Group” at INESC-ID. I investigate the creation of AI and complex systems using an agent-based approach, with a special focus on social agents. My main research interests are in the fields of Autonomous Agents and Multi-Agent Systems, Affective Computing, Virtual Agents and Human-Robot Interaction.

 

The world is changing! Intelligent and autonomous machines are entering not only our workplace but also our homes. And as such, machines must be social. I believe that one of the challenges that the areas of Artificial Intelligence (AI) and Robotics face nowadays is how to make machines “social”, that is, how to endow agents and robots with  social interaction capabilities enabling them to interact with humans and with other agents in a natural, flexible and transparent way.

My main research focuses on the problems and techniques for creating social agents that can simulate human-like behaviours, be transparent, natural and eventually, give the illusion of life. Over the years I’ve dealt with this problem by engineering agents that exhibit specific social capabilities, including aspects such as emotions, personality, culture, non-verbal behaviour, empathy, collaboration, and others.  My main contributions to the area of social agents have been in the field of virtual agents, multi-agent systems, affective computing and social robotics. Most of my publications on these topics can be found in the GAIPS web site, or through my google scholar.

Studying social interactions “for” and “with” machines is a challenge that I believe will allow us to engineer engaging, natural, and, most importantly, “humane” AI. That is my journey…

Research

Most of the research I’ve developed over the years has focused on the affective and social aspects of agents and how they contribute to achieving social intelligent machines.  I believe that to achieve intelligent machines we need to build mechanisms and develop techniques that endow machines of “social competencies”. That has been my vision over the years, and my research was conducted through collaborative work with many partners all over the world, and through the participation in several national and international projects in the areas of social robotics, virtual agents, affective computing and serious games.  Some of these international projects  (eg. the LIREC, ECUTE, HUMAINE, EMOTE, AMIGOS projects) have influenced me deeply and shaped what I believe the area of Social Artificial Intelligence should be.

Further, I’ve been very lucky to be among the few set of academics that were at the forefront of innovative areas such as Intelligent Intelligent Virtual Agents and Affective Computing, having witnessed their birth, and how they are nowadays changing what we believe AI and computing is.

The areas that I’ve been working over the years are:

Pro-social Computing

Pro-social computing is one if my current challenges!

I’m working towards a future where autonomous agents are used to fostering and supporting pro-social behaviour in a hybrid society of humans and machines. Pro-social behaviour occurs when people and agents perform costly actions that benefit others. Acts such as helping others voluntarily, donating to charity, providing information or sharing resources, are all forms of pro-social behaviour. Two main questions guide this research which challenges the purely utilitarian view of human decision making and its role in hybrid societies: What are the conditions and mechanisms that lead societies of agents and humans to be more pro-social? How can we engineer autonomous entities (agents and robots) that lead to more altruistic and cooperative behaviours in a hybrid society? These are some of the questions that I’ve been trying to address in my research in the area of social agents.

See more in the AAAI’18 paper.

News

2018

  • My student, Elmira Yadollahi (from the IST-EPFL PhD program and co-advised by Pierre Dillenbourg) won the “Student ICC award from ACM” at the IDC (Interaction Design for Children) conference this year with her work “When deictic gestures in a robot can harm child-robot collaboration”. This work also won an honorable mention in the best paper award category in IDC. I’m so proud of her!
  • Hang Yin, one of my PhD students doing a combined PhD program with EPFL, co-supervised with Prof. Aude Billard and Prof. Francisco Melo, defended his thesis on the 5th of June. Congratulations Hang!
  • I will be giving a keynote talk at the next AAMAS conference (Autonomous Agents and Multi-agent Conference) in Sweden, next July, with a title “Ready Team Player One: Social Robots in Teams”.
  • I will be giving a keynote talk at the next UMAP conference (User Modeling Adaptation and Personalisation) in Singapore, next July, with a title “Robots that listen to people’s hearts: the role of emotions in the communication between humans and social robots”.
  • I gave an invited talk on “The role of emotions in the communication between humans and social robots“, at the workshop on Affective artificial agents as models for affective science and psychology, organised by the Swiss Center for Affective Sciences, Geneva, April 9th, 2018.
  • Our paper “Engineering Pro-sociality with Autonomous Agents” (co-authored with Francisco C. Santos and Fernando Santos), presented at  AAAI’18 this year, won the first prize of the Blue Sky Awards sponsored by the Computing Community Consortium (CCC) at the AAAI-18 Senior Member Presentation Track.
  • Our paper in “Justifying Failures in Human-Robot interactions” has been accepted as a full paper for the AAMAS’18!
  • The GAIPS group just got three papers accepted at HRI’18 in Chicago in March. Two of the papers are associated with the AMIGOS project and one with the Co-Writer project. We are so happy! Chicago, here we go..