Abstract for AISB 2000: How to Design a Functioning Mind

Abstract for the
Symposium on How to Design a Functioning Mind
17-18th April 2000
At the AISB'00 Convention


AUTHOR: Frederic ALEXANDRE             Tel: (+33/0) 3 83 59 20 53
INRIA-Lorraine/LORIA-CNRS              Fax: (+33/0) 3 83 41 30 79
BP 239                                 E-mail: falex@loria.fr
54506 Vandoeuvre-les-Nancy Cedex       http://www.loria.fr/~falex
FRANCE

POSTER TITLE: Inspiration from Neurosciences to emulate
    Cognitive Tasks at different Levels of Time

ABSTRACT:
Our team has been working for more than ten years on the
modelling of biologically inspired artificial neural networks.
Today, our models are used to different cognitive tasks like
autonomous behavior and exploration for a robot, planning,
reasoning, and other tasks linked to memory and internal
representation building. We propose to present the framework that
underlies these models through the time delays related to several
fundamental properties like information coding, learning,
planning, motivation.

More precisely, at the level of the spike, we will describe how
population of neurons can synchronize to perform elementary
functions. At the level of information transmission in  cortical
maps, we will present how mapping can be built and exchange
information. At the learning level, we will propose models for
procedural, semantic, episodic and working memory. At the
behavioral level, we will present real-world tasks, integrating
the above mentionned levels, endowing an autonomous robot with
such capacities as exploration, cognitive map building, planning,
motivation management.

In a compact way (because of poster format), emphasis will be
given to relations between biological data, computer science
model and real-world implementation.

================================================================
4. Short CV:

I am a Senior Research Scientist of INRIA at
INRIA-Lorraine/LORIA-CNRS in Nancy, France. I hold a B.S. in
Computer Science (1986) from the Institut National Polytechnique
de Lorraine and a PhD in Computer Science from the University
Henri Poincare', Nancy, where I defended my dissertation (1990)
entitled "A functional modelization of the cortex: the cortical
column". I also have a Master's degree in Psychology (1992), one
in Physiology (1991) and another in Mechanics (1985). In 1997, I
defended my habilitation to be a director of research entitled
"Neuromimetic Intelligence" at the University Henri Poincare',
Nancy, France.

My current research interests concern neurosymbolic integration
with special focus on unified connectionist models. They lead to
biologically inspired architectures and learning rules for
neuronal network design with application to associative learning,
vision, language, reasoning and autonomous behavior. As the
leader of the CORTEX team (15 people) which concentrates
connectionist activities in INRIA-Lorraine/LORIA-CNRS, I have
developed and refined a number of biologically plausible
connectionist mechanisms and have participated in industrial
applications involving the use of connectionist and symbolic
tools in integrated hybrid systems. I was the supervisor of
research for 10 PhD students (and currently for 5).

I am the author of more than 100 papers (including the edition of
one book and 18 papers in journals). I have co-chaired the IJCAI
Workshop on Connectionist-Symbolic Integration, 1995 and
organised (or participated to the organisation of) several
conferences about artificial neural networks. I was the leader
the european ESPRIT project MIX on modular integration of
connectionist and symbolic processing in knowledge-based systems
and my team also belongs to the european networks of excellence
NEuroNet and NeuroColt.