Temple University, USA
Special Title:
Optimization of
Federated Learning
at Edge Clouds:
Participant
Selection and
Learning Scheduling
Abstract:
Edge computing and
federated learning
(FL) have gained
popularity since
they provide a
promising edge
learning framework
that mitigates the
limitations of long
latency, high cost,
and privacy concerns
in cloud-based
centralized
learning. While most
existing works on
federated learning
over edge systems
focus on optimizing
the training of the
global model in edge
systems, the
concurrent training
of multiple FL
models from
different
applications in a
shared edge cloud
can lead to edge
resource competition
and affect the
training performance
of each model.
Hence, in this talk,
I will present our
recent works in
addressing this
challenge by
proposing
optimization
algorithms to
jointly select FL
participants and
learning rates or
topologies for each
model, with an aim
of minimizing the
total training cost.
I will first
introduce a
multi-stage
optimization
framework that
allows FL models to
select their
participants and
learning rates or
learning topologies.
Then, I will
describe a quantum
assisted algorithm
to tackle the joint
participant
selection and
learning scheduling
problem using both
quantum and
classical computing.
Last, I will briefly
discuss our proposed
solution for group
formation and group
sampling for
group-based FL over
edge systems.
Biodata:
Yu Wang is a
Professor in the
Department of
Computer and
Information Sciences
at Temple
University. He holds
a Ph.D. from
Illinois Institute
of Technology, an
MEng and a BEng from
Tsinghua University,
all in Computer
Science. His
research interest
includes wireless
networks, smart
sensing, and mobile
computing. He is a
recipient of Ralph
E. Powe Junior
Faculty Enhancement
Awards from Oak
Ridge Associated
Universities (2006),
Outstanding Faculty
Research Award from
University of North
Carolina at
Charlotte (2008),
Fellow of IEEE
(2018), and ACM
Distinguished Member
(2020). He has
served as Associate
Editor for IEEE
Transactions on
Parallel and
Distributed Systems,
IEEE Transactions on
Cloud Computing,
among others.
Prof. Makoto Iwasaki (IEEE Fellow)
Nagoya Institute of Technology, Japan
Speech Title:
GA-Based Practical System Identification and Auto-Tuning for Multi-Axis Industrial Robots
Abstract:
Fast-response and
high-precision
motion control is
one of indispensable
techniques in a wide
variety of high
performance
mechatronic systems
including micro
and/or nano scale
motion, such as data
storage devices,
machine tools,
manufacturing tools
for electronics
components, and
industrial robots,
from the standpoints
of high
productivity, high
quality of products,
and total cost
reduction. In those
applications, the
required
specifications in
the motion
performance, e.g.
response/settling
time,
trajectory/settling
accuracy, etc.,
should be
sufficiently
achieved. In
addition, the
robustness against
disturbances and/or
uncertainties, the
mechanical vibration
suppression, and the
adaptation
capability against
variations in
mechanisms should be
essential properties
to be provided in
the performance.
The keynote speech
presents a practical
auto-tuning
technique based on a
genetic algorithm
(GA) for servo
controllers of
multi-axis
industrial robots.
Compared to
conventional manual
tuning techniques,
the auto-tuning
technique can save
the time and cost of
controller tuning by
skilled engineers,
reduce performance
deviation among
products, and
achieve higher
control performance.
The technique
consists of two main
processes: one is an
autonomous system
identification
process, involving
the use of actual
motion profiles of a
typical robot. The
other is an
autonomous control
gain tuning process
in the frequency and
time domains,
involving the use of
GA, which satisfies
the required tuning
control
specifications,
e.g., control
performance,
execution time,
stability, and
practical
applicability in
industries. The
proposed technique
has been practically
evaluated through
experiments
performed with an
actual six-axis
industrial robot.
Biodata:
Makoto Iwasaki
received the B.S.,
M.S., and Dr. Eng.
degrees in
electrical and
computer engineering
from Nagoya
Institute of
Technology, Nagoya,
Japan, in 1986,
1988, and 1991,
respectively. He is
currently a
Professor at the
Department of
Electrical and
Mechanical
Engineering, Nagoya
Institute of
Technology.
As professional
contributions of the
IEEE, he has
participated in
various organizing
services, such as, a
Co-Editors-in-Chief
for IEEE
Transactions on
Industrial
Electronics from
2016 to 2022, a Vice
President for
Planning and
Development in term
of 2018 to 2021,
etc. He is IEEE
fellow class 2015
for "contributions
to fast and precise
positioning in
motion controller
design".
He has received many
academic,
foundation, and
government awards,
like the Best Paper
and Technical Awards
of IEE Japan, the
Nagamori Award, the
Ichimura Prize, and
the Commendation for
Science and
Technology by the
Japanese Minister of
Education,
respectively. He is
also a fellow of IEE
Japan, and a member
of Science Council
of Japan.
His current research
interests are the
applications of
control theories to
linear/nonlinear
modeling and
precision
positioning, through
various
collaborative
research activities
with industries.
Speakers in 2024 to
be announced
soon......