Place: Room 621/622, GP
South (Building 78)
Time: Thursday 29th
November, 10:30 morning Tea, 11:00am seminar
Title: In-situ Distributed
Genetic Programming for On-line Learning in
Complex Systems
Speaker: Philip Valencia,
CSIRO ICT
Host: Prof Peter Lindsay
Abstract:
In Nature, evolution utilises energy to
both optimise the current olution in the
current environment as well as exploring
innovative solutions for future
unknown environments. Over many
entity lifetimes, multi-agent systems can
evolve to exhibit intelligent
collective behaviours such as that seen with
bees and ants. The collective behaviour is a
result of the co-evolution of the
agents' embodiments as well as their on-line
adaptation and learning capacity (i.e.
brains). Interestingly higher-order systems
can evolve logic solutions to many problems
within the lifetime of the agent, despite a
fixed embodiment. Such exploration and
optimsation of logic can be described as
"playing" or "training", while
the exploitation of logic is akin to "work".
For these concepts to even
exist, a number of specific conditions must
be present which will be discussed in this
presentation.
This research conjectures that agents within
a complex real world environment can perform
genetic evolution of logic to
yield desirable and adaptive collective
system behaviours. Typically the logic
for man-made complex systems is devised
off-line, centralised and biased by human
intuition and simulation accuracy. It will
be
argued that such solutions are inevitably
"brittle" and that the in-situ evolution of
logic is preferable for specific
problem domains.
Biography:
Philip Valencia received his Bachelors
degrees in Engineering (Electronic) and
Information Technology from QUT, Australia
in 2001. He worked at the CSIRO in Sydney
researching real-time time facial
recognition and machine learning before his
involvement in the Ageless Aerospace Vehicle
(AAV) project - a NASA funded project
looking at futuristic aerospace vehicles.
For the next 3 years Philip's research
focussed on Multi-Agent System (MAS) and
Complex Systems Science as methods for
achieving self-diagnosing and self-repairing
structures. Through this Philip became
involved in distributed sensing and
actuation for environmental resource
management before returning to Brisbane as a
research engineer for the newly formed CSIRO
ICT Centre (Autonomous Systems Laboratory).
Here he continued researched on Wireless
Sensor Networks (WSN).
Dr. Ariel Liebman
Research Fellow
ARC Centre for Complex Systems
School of ITEE, University of Queensland
email:
a.liebman@uq.edu.au
tel: +61-7-3365-1623
mobile: +61-(0)414-226-336
room: 78-414, GP South, St Lucia Campus
web: www.accs.edu.au