Wednesday, April 23, 2014 10:00 am
-
10:00 am
EDT (GMT -04:00)
MC 6496
Speaker
Zhen Wang, Department of Applied Mathematics, University of Waterloo
Title
Clustering Behavior in Neural Networks with Time Delay
Abstract
A
neural
network
is
an
information
processing
paradigm
that
is
inspired
by
the
way
biological
nervous
systems,
such
as
the
brain,
process
information.
It
is
composed
of
a
large
number
of
highly
interconnected
processing
elements
(neurons)
working
in
unison
to
solve
specific
problems.
It
is
well
known
that
oscillations
occur
in
many
neural
networks
models,
and
the
properties
of
the
oscillations
depend
on
the
characteristics
of
the
individual
neurons,
how
the
neurons
are
connected
to
each
other
and
the
presence
of
time
delays
in
the
connection.
Cluster
state
is
a
special
state
of
the
oscillations
in
which
multiple
subpopulations
coexist
and
each
of
the
cluster
consists
of
fully
phase
synchronized
oscillators.
In
this
talk,
I
will
give
a
brief
review
of
modelling
of
neural
networks
and
literature
in
this
field,
then
introduce
the
linear
stability
results
about
a
system
of
N
globally
coupled
oscillators.
Finally,
an
example
model
for
six
globally
coupled
Morris-Lecar
oscillators
is
presented
to
show
how
to
use
phase
model
to
analyze
the
stability
of
clustering
behavior.