Ruth Aylett is a Professor of Computer Science at Heriot-Watt University in Edinburgh and a member of the Edinburgh Centre for Robotics. She was one of the founders of IVA and has worked in graphical agents as well as social robotics for more than 20 years. Projects include an anti-bullying agent, an agent for improving cultural sensitivity, long-lived robot companions, and an empathic robot tutor. She has just completed a project on the use of a robot to improve social signal detection for high-functioning adults with autism and has just started a project to create a story-telling agent for reminiscence therapy with adults with Alzheimer’s. She is interested in the links between affective architectures and external behavior in both graphical and robotic agents.
Keynote: Expressive behavior, social signals, and embodiment
In this talk, I look at the relationship between expressive behavior for an embodied character (graphical or robotic), and its social signals. I argue that where an embodied character runs a cognitive architecture, there is a tendency for naïve production of expressive behavior as a straight representation of its internal state. However, in humans, this is rarely the case and expressive behavior is usually at the service of social signaling. I propose that strong embodiment, in which bodily processes are modeled, can contribute to a more sophisticated account and provide extra capabilities such as micro-expressions and physiological states such as blushing and sweating.
Claudio Pinhanez is a scientist, innovator, and professor. He leads research in Conversational Intelligence in the laboratory of IBM Research in Brazil. He is also the Deputy Director of the C4AI, the Center for Artificial Intelligence created by a partnership between the University of São Paulo, IBM, and FAPESP. Claudio is an expert in artificial intelligence, human-machine interaction, conversational systems, and service science. He has worked with AI since he got his Ph.D. at the MIT Media Laboratory in 1999 and has more than 150 papers published in journals and scientific conferences, and more than 30 patents issued in the USA, Europe, and Japan.
Keynote: New Challenges and Opportunities in Conversational Agents
This talk addresses new challenges and opportunities in conversational agents after a decade of tremendous scientific progress and commercial success. I will explore some of the recent advances in Foundation Models and how they are bringing new challenges and opportunities for the design and development of conversational agents. I will review some recent results of my research regarding the use of formal language and gender biases and use them to discuss some key foundational and ethical issues in the context of agents for low-resource and endangered languages.