UNIVERSITY OF HERTFORDSHIRE
             COMPUTER SCIENCE RESEARCH COLLOQUIUM

                            presents

         "From Information-Theoretic Behaviour Generation
               towards Agent-Agent Interaction"


                        Christoph Salge
                 (University of Hertfordshire, UK)

                    9 March 2011  (Wednesday)
                      Meeting Room LD454

                  Hatfield, College Lane Campus
                            3 -4  pm

                  Everyone is Welcome to Attend

                  Refreshments will be available

Abstract:

To improve our understanding of social interaction in a developmental
context it is important to know what the benefit of social interaction
is.  Once a collective of individuals is observed to exhibit social
interaction, a post hoc analysis and explanation of its advantages
is often straightforward. From an evolutionary or developmental point
of view, however, the challenge is to identify a path that would lead
towards such an interaction in the first place.

We start with the assumption of a single agent that generates its
behaviour based on the driving forces of information maximization and
information parsimony. This work is based on the recent advances in
information theory based behaviour generation, especially Vergassola's
idea of infotaxis.

Within a low-assumption agent-world interaction framework, based on
Information Theory and Causal Bayesian Networks, we then demonstrate how
every agent that needs to acquire relevant information in regard to its
strategy selection will automatically inject part of this information back
into the environment. We introduce the concept of 'Digested Information'
which both quantifies, and explains this phenomenon. Based on the
properties of digested information, especially the high density of
relevant information in other agents actions, we outline how this could
motivate the development of low level social interaction mechanisms,
such as the ability to detect other agents.

We also examine the emerging phenomena in information based agent-agent
interaction. One interesting result is the replication of the three
boids rules for flocking behaviour (Alignment, Cohesion, Separation),
providing a possible deeper explanatory level for flocking behavior.

---------------------------------------------------
Hertfordshire Computer Science Research Colloquium
http://homepages.stca.herts.ac.uk/~nehaniv/colloq