CALL FOR PAPERS

AGENTS IN INTERACTION - ACQUIRING COMPETENCE THROUGH IMITATION Full-day Workshop, May 10, 1998

Associated with Second International Conference on  AUTONOMOUS AGENTS (Agents '98)
Minneapolis/St. Paul, May 10-13, 1998

Co-organisers:  Kerstin Dautenhahn* and Gillian Hayes**

*Department of Cybernetics, University of Reading, UK
**Department of Artificial Intelligence, University of Edinburgh, UK

     http://cyber.reading.ac.uk/staff/people/kd/WWW/aaimi.html

The scope of the workshop encompasses social learning and imitation as a means of one agent, software or embodied, learning an individual behaviour pattern or utterance from a member of the same or a different species and including it in its own behavioural repertoire. The workshop is intended to attract people from different communities where social learning and imitation is involved, i.e. where agents learn from each other or their users through interaction.
 

AREAS OF INTEREST

IMITATION AND SOCIAL LEARNING IN AGENTS

Imitation is supposed to be among the least common and most complex  forms of animal learning. It is found in highly socially living  species which show, from a human observer point of view, 'intelligent'  behaviour and signs for the evolution of traditions and culture. There  is strong evidence for imitation in certain primates (humans and  chimpanzees), cetaceans (whales and dolphins) and specific birds like  parrots. Recently, imitation has begun to be studied in domains  dealing with such non-natural agents as robots, as a tool for easing  the programming of complex tasks or endowing groups of robots with the  ability to share skills without the intervention of a  programmer. Imitation plays an important role in the more general  context of interaction and collaboration between agents and humans,  e.g. between software agents and human users. Intelligent software  agents need to get to know their users in order to assist them and do  their work on behalf of humans. Imitation is therefore a means of  establishing a `social relationship' and learning about the actions of  the user, in order include them into the agent's own behavioural  repertoire.

The role of imitation as an effective learning mechanism is important  in engineering domains. The main aims are to develop imitation as a  machine learning method for an agent, allowing it to learn from one or  very few examples which are performed by a model, and to facilitate  indirect knowledge transfer from one agent to another. The latter  becomes more and more interesting for scenarios where interactions  between heterogeneous agents are studied, because in these situations  the simple transfer of a successful control program from one agent to  another is often impossible because of great differences in  construction and behavior characteristics.  Examples include the use  of imitation of movements by a robot to learn a navigation task, and the acquisition of a synthetic robotic language by observation.  The  obvious extension is to situations where the agent has to learn from a  human `model' in, for example, the context of service robots which  must adapt to humans and cooperate/work hand-in-hand together with  humans.  In Artificial Life research on individualized robot  societies, imitation is used as a social mechanism for identifying and  building up social relationships towards robot group members. In  Software Agent research, imitation is used as a means of enabling  agents to adapt to one another and develop a coherent group behaviour.

As a research topic imitation tackles such fundamental problems as  sensory intelligence, motor control, real-time learning architectures,  intermodal representation, social interactions, motivational and  emotional control of behavior, and scaling-up from sensorimotor  intelligence to cognitive systems. Generally, different mechanisms are  studied in these different domains, so the problem arises of  integrating them in a common framework.  The topic of imitation is  broad enough to cover all these interesting issues.

The aim of the workshop is to draw together researchers working in  software, hardware and wetware fields with the common goal of  understanding the role of social learning in making agents useful,  believable, acceptable or simply natural.
 

WORKSHOP FORMAT

The Workshop will comprise a few keynote talks, a panel discussion  with participants from different research areas, and sessions with  presentation of state-of-the-art social learning and imitation  research.
 

SUBMISSION DETAILS

People who are interested in participating in the workshop are asked  to submit an extended abstract (not more than 4 pages).  Please submit  4 hardcopies to:

Kerstin Dautenhahn
Department of Cybernetics
The University of Reading
Whiteknights, PO Box 225
Reading, RG6 6AY, United Kingdom
tel: +44 (0) 118 931-8219 or -6372
fax: +44 (0) 118 931-8220
K.Dautenhahn@cyber.reading.ac.uk

An email submission in plain ascii format is also possible, postscript  submissions cannot be accepted.
 

IMPORTANT DATES

January 15, 1998: Workshop papers due
February 28, 1998:  Notification
March 30, 1998: Final Copies for Workshop Notes Due
 

PROGRAM COMMITTEE

Kerstin Dautenhahn (University of Reading, UK)
Gillian Hayes (University of Edinburgh, UK)
Roy Middleton (Edinburgh Virtual Environment Centre, UK)
Peter McOwan (University of Reading, UK)
Simon Penny (CMU, USA)
Paolo Petta (Austrian Research Institute for Artificial Intelligence)
Angi Voss (GMD, Germany)
 
 
Last Modified on