Abstract
Training of interpersonal communication skills is typically done using role play, by practising relevant scenarios with the help of professional actors. However, as a result of the rapid developments in human-computer interaction, there has been an increasing interest in the use of computers for training of social and communicative skills. This type of training offers interesting opportunities to complement traditional training methods with a novel paradigm that is more scalable and cost-effective. The main idea underlying such applications is that of a simulated conversation between a human trainee and an artificial interlocutor (or 'virtual agent') represented in a virtual environment (e.g., using a tablet, laptop, or Virtual Reality headset). By developing the system in such a way that the (verbal and/or non-verbal) communicative behaviour of the human has a direct impact on the behaviour of the virtual agent, an interactive learning experience is created. In this talk, I will discuss the current state-of-the-art in virtual agent-based communication training. I will provide an overview of the popular application domains, review various interaction modalities that can be used, and discuss the challenges involved in creating a realistic user experience and evaluating the effectiveness of training systems. Most of these issues will be illustrated based on my group's recent work on aggression de-escalation training.
Speaker Bio
Tibor Bosse is Full Professor of Artificial Intelligence and Communication Science at the Communication & Media group at Radboud University, where he is currently the Director of Education. Moreover, he is Scientific Director of the OnePlanet ICAI Lab For Precision Health, Nutrition & Behaviour, Member of the Strategy Team of the Netherlands AI Coalition, Chair of the NeFCA working group on Human Machine Communication, Chair of the UvN Expert Table Artificial Intelligence, Figurehead of the NWA Big Data Route and former President of the Benelux Association for Artificial Intelligence. In his research, he builds bridges between Computer Science and the Social Sciences, in particular between Artificial Intelligence and Communication Science. In particular, he combines insights from these disciplines in order to design, implement and evaluate ‘Social AI’ systems, i.e. intelligent computer systems that have the ability to engage in natural social interactions with human beings. Such systems include, among others, social robots, virtual agents and chatbots. From a technical perspective, Bosse develops new algorithms to endow social AI systems with more human-like behaviour (e.g., using techniques such as emotion recognition and natural language processing). From a social perspective, he studies the psychological effects of interacting with social AI systems on people’s perceptions and behavior. Finally, from an applied perspective, he is interested in the use of social AI systems for various practical purposes, including social skills training and behavior change.
0 Praat mee