Position statement on agent standardisation

Caveat

This is a personal position statement and the views expressed herein do not necessarily reflect those of Sharp Laboratories of Europe Ltd. or of Sharp Corporation.

Background

In the Distributed Intelligent Systems team at Sharp Laboratories of Europe we have been conducting multi-agent research for several years. During this time we have built several distributed prototype / demonstrator systems using tools developed in-house. Our approach is based on static knowledge-based agents communicating via speech act messages. The internal architecture of our agents is based loosely on Shoham's Agent0 work. Our speech act categories are derived from Searle's original illocutionary acts.

Application area

Our view of a personal agent is as software resident on item of personal electronic equipment which acts as an intermediary between the user and the complexities of electronic services in the widest sense. The purpose of the agent is to provide extra value, flexibility and functionality to the user while at the same time simplifying operation. The personal agent knows the characteristics, preferences and habits of its user through a combination of direct instruction, learning by observation and reinforcement learning. It also knows about information and other services available across the network, and how to exploit them on behalf of its user, again, through a combination of instruction and observation. The agent can act autonomously, for example acquiring a document it believes its user will want to view and bring it to the user' attention. It can also carry out instructions, interpreting high-level, user-oriented commands in terms of more specific actions which will bring about the desired results.

Viewpoint on standardisation

My view is that agents will be used to provide an extra layer of "control" to augment existing and projected multimedia content delivery services working over public networks. The agents will exchange information on the availability of services and content, negotiate over terms and conditions, etc., then when agreement is reached, authorise delivery. Examples of delivery services include:

• Internet/HTTP
• interactive digital broadcast, VOD, etc.
• telephony-related services

I would like to see FIPA tackle definition of a message language and message delivery service for this purpose. The various levels of language requiring definition include:

1. speech act level (cf KQML)
2. grammar level (cf KIF)
3. domain ontology
4. negotiation protocols

My view is that we should aim to define a simple, practical, but extandable language rather than a complex one which aims to solve all problems in one go.

Although the above is phrased in static agent / speech act terms, an equivalent set of requirements could be drawn up for mobile agents. Perhaps FIPA should aim to support both in an integrated fashion.

Paul Kearney
Research Manager, Distributed Intelligent Systems
paul.kearney@sharp.co.uk
Sharp Laboratories of Europe Ltd.
Oxford Science Park
Oxford OX4 4GA