►
From YouTube: What, Why and How of Federated Machine Learning – Implementation Using... Neeraj Arora & Layne Peng
Description
Don’t miss out! Join us at our next event: KubeCon + CloudNativeCon Europe 2022 in Valencia, Spain from May 17-20. Learn more at https://kubecon.io The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy, and all of the other CNCF-hosted projects.
What, Why and How of Federated Machine Learning – Implementation Using FATE, KubeFATE, and FATE-Operator for Kubeflow - Neeraj Arora & Layne Peng, VMware
Federated Machine Learning (FML) is an emerging technology that makes it possible to extract insights from widely dispersed data sources. It may also be used with privacy enhancing measures to reveal insights without revealing the underlying data. In doing so, it gives global organizations an opportunity to utilize worldwide data while adhering to regulations for different geographic regions, allows insights to be developed without moving large amounts of data to a central location, and in a multi-party scenario allows each party to control their individual decision to participate. This talk will introduce FML discussing its basic principles, its manifestations, high-level use-cases, and tie these in with the technology being developed. To move from theory to practice, we will demonstrate how FML can be implemented on Kubernetes using FATE, KubeFATE, and integrated with Kubeflow using the FATE-Operator. We’ll also discuss new features being implemented into FATE for realistic use-cases. A similar presentation was delivered at CNOS Virtual Summit China in July 2020 and focused on the integration of FATE into Kubeflow. The current presentation will cover technical ground instead on the challenges and opportunities of FML for real-world use-cases encountered since.
A
We
are
also
involved
laying
more
deeply
than
I
in
fate,
which
is
one
of
the
many
ways
you
can
realize.
Federated
ai
fate
is
an
open
source
project
and
is
discussed
in
this
presentation
in
this
session.
We'll
cover
federated
machine
learning
at
a
conceptual
level.
Before
looking
at
fate
broadly
followed
by
a
demo,
we
have
been
working
on
the
fml
site
portal
and
lane
will
take
you
through
a
demonstration
of
that.
Also.
A
A
B
A
Want
to
use
are
available
via
libraries.
Computing
power
has
also
been
getting
cheaper
and
readily
accessible.
The
barrier
might
be
cost,
but
for
the
most
part
it
is
readily
available.
Commercially
data,
however,
is
a
different
story
due
to
regulations
or
sheer
logistics,
or
the
foresight
required
to
model
a
problem
and
the
cost
associated
with
collecting
a
large
quantity
of
data.
Getting
the
right
and
enough
data
for
model
building
remains
done
as
a
community
and
industry.
A
These
are
the
type
of
problems
that
federated
learning
can
help
with.
There
are
a
couple
of
excellent
resources
by
others,
specifically
google,
that
can
help
explain
the
idea
in
detail
and
I'll
try
and
introduce
it
simply
in
what
we
might
call
traditional
machine
learning.
We
bring
all
the
data
together
in
one
central
location
where
one
or
many
compute
resources,
as
might
be
the
case
with
distributed
clustered
systems,
train
a
model
using
the
entirety
of
the
data
in
a
federated
learning
system.
A
A
Significant
amount
of
work
has
been
done
and
is
ongoing
since
federated
learning
was
first
proposed.
This
is
an
excellent
survey
paper
if
you're
interested
as
you
read
through
it,
you'll
see
that
federated
learning
provides
some
very
novel
answers
to
the
problems
with
data
science.
It
is
not
difficult
to
imagine
how
federated
learning
may
affect
and
change
the
world
of
iot
and
industry.
Ledge
I'll,
try
and
illustrate
the
essential
concept
of
federated
learning
graphically
and
we
use
federated
learning
and
federated
machine
learning
interchangeably
in
this
presentation.
A
A
The
nodes
train
the
model
they
receive
using
their
local
data
and
send
the
updated
model
back
to
the
aggregator
which
combines
them
into
an
updated
model.
Surely
I
have
simplified
the
process
for
explanation
instead
of
the
entire
model
being
exchanged.
For
example,
only
the
delta
weights
need
to
be
exchanged,
but
the
concept
is
the
same.
If
you
will,
let's
look
at
the
process
again,
but
this
time,
let's
extend
our
imagination
to
the
three
phases
employed
in
model
building,
namely
training,
validation
and
testing.
A
Not
all
nodes
always
participate
in
training
or
validation
of
a
federated
model.
The
aggregator
or
the
server
decides
which
nodes
participate
in
a
run
or
an
epoch
by
excluding
participants
and
consequently
the
data
those
participants
hold
locally.
We
are
dividing
the
data
set
into
subsets.
Just
like
we
divide
data
in
a
data
set
into
training,
sets,
validation,
sets
and
testing
sets
nodes
that
participate
in
training
in
an
epoch
will
be
excluded
from
validation
and
testing
in
that
epoch.
A
Overall,
there
are
three
types
of
federated
machine
learning
at
the
moment:
there's
horizontal
federated
learning,
vertical
federated
learning
and
federated
transfer.
Learning.
Horizontal
learning
is
where
the
same
fields
exist
with
multiple
nodes.
For
example,
branches
of
a
bank
that
use
the
same
software
may
store
the
same
fields
about
completely
different
users.
A
A
Vertical
learning
is
where
different
fields
exist
with
multiple
nodes,
regardless
of
the
entries
in
the
data
having
any
relationship
or
not.
A
good
example
of
this
would
be
different
merchants
or
businesses
that
service
the
same
population,
such
as
a
credit
card
issuer
like
a
bank
that
issues
credit
cards
which
customers
use
to
pay
utilities.
A
Using
that
information
to
tailor
capacity
for
changing
demands,
federated
transfer
learning
is
applicable,
similar
to
regular
transfer
learning
where
generalized
models
with
aggregators
can
be
specialized
for
local
use
by
node,
for
example,
infectious
disease
detection
at
hospitals,
fraud,
detection
across
banks
operating
in
different
geographies
and
fault,
and
accident
prevention
by
mechanical
failure
across
an
entire
industry
that
uses
similar
equipment
summarizing.
Simply
federated
learning
attempts
to
address
some
of
the
issues
that
we
are
seeing
more
and
more,
namely
the
issue
of
privacy,
trust
enforced
either
by
user
choice
or
by
regulations.
A
The
issue
of
aggregating
lots
of
data
in
one
location,
the
lack
of
data
itself,
possibly
driven
by
the
cost
of
bringing
significant
amount
of
data
together
depending
on
the
kind
of
nodes
you
deploy.
It
may
help
with
sourcing
compute
by
conquering
a
large
problem
by
not
only
dividing
the
data
but
also
the
computational
load
and
depending
on
your
implementation,
it
can
be
highly
scalable.
A
Although
federated
learning
brings
in
these
benefits,
it
also
presents
some
challenges
being
a
heavily
distributed
system.
It
can
get
tougher
to
secure
data
and
apply
security
policies.
This
could
compromise
data
confidentiality
with
emergent
attack,
vectors
being
distributed
will
also
affect
availability
of
both
compute
and
data.
There's
also
the
issue-
and
this
depends
on
your
use
case,
of
the
integrity
of
data
that
remote
nodes
are
presenting
at
a
basic.
B
A
Infrastructure
events
that
are
relatively
easily
solved
in
a
data
center
environment
become
complex
in
a
distributed
network
and,
more
so
in
a
widely
distributed
network.
These
are
some
things
you
want
to
keep
in
mind.
As
you
consider
federated
hitler,
we
saw
earlier
the
three
pillars
that
support
ai
and
machine
learning.
A
I'd
like
to
add
a
fourth
pillar
here,
ops,
it
is
one
thing
to
build
a
model,
but
entirely
another
to
include
it
in
a
product.
Models
are
also
often
refined.
With
new
data
that
is
constantly
collected,
federated
ai
technology,
enabler
or
fit
is
a
suite
of
libraries
and
services
you
can
deploy
in
various
forms
on
various
form
factors
to
enable
your
solutions
with
federated
learning.
A
A
The
ops
pillar
is
critical
for
success
with
any
ml
or
ai
initiative.
In
production,
our
work
on
fate
adds
in
pluggable,
compute
libraries,
data
sources,
the
ability
to
scale
across
clusters
and
nodes,
full
life
cycle
model,
management
of
fml
generated
models
with
resiliency
against
infrastructure
events
being
built
in
with
the
right
combination
of
third-party
products.
You
can
also
extend
the
capability
of
fate.
B
Thank
you.
I
am
name
time.
I
will
cover
the
projects
we
contribute
and
working
on
to
bring
family
learning
into
production
easier,
and
then
you
demo
a
full
flow
of
for
federation
learning
case.
We
know,
besides
the
complexity
of
normal
machine
learning.
Factor
learning
brings
actual
operational
challenges
to
cluster
management,
to
how
easy
to
manage
the
fair
learning
cluster
with
initial
and
open
source,
the
project
kill
fade,
which
is
part
of
a
fade
community,
to
manage
the
fair
running
cluster
in
one
site:
the
cubase,
decompose
and
kubernetes
environment.
B
B
B
When
we
meet
real
users
from
transitional
industry,
there
is
a
strong
requirement
to
lower
the
bank
to
use
fair
learning.
Photography,
fake
design
as
and
fade
is
design
and
interlock
timeline,
and
with
a
lot
of
outposts
federal
learning
operation,
we
can
program
our
application
as
composing
a
pipeline.
B
The
fml
side
portal
provides
federated
site
management,
data
management,
model
management
and
project
management
features
in
a
web,
gui
style
with
it.
We
can
upload
and
engineering
the
data
manager
data
contributes
to
the
fair
learning
paper
outbound
operations
in
faith
such
as
homogeneous,
logistic
regression,
homogeneous
reversed,
etc
and
rating
the
result
all
these
by
step
and
put
them
into
the
pipeline
and
kickstart
the
fair
running
job
or
by
clicking
the
mouse
for
easier
to
get
what
I
introduced
the
face
to
face
fmsi
portal.
Let
me
show
a
quick
demo
in
this
demo.
B
B
We
first
take
code
application
in
a
triple
role
book,
and
then
we
will
show
the
gui
introduction
of
the
xml
site
portal.
In
this
demo,
we
use
open
source
cache
data
credit
card
flow
detection,
as
data
set
as
we
prepared
a
small
cognitive
cluster
to
simulate
a
rate
management
side
and
other
two
collected
cluster.
B
B
B
B
Then
we
use
the
jupyter
notebook,
which
is
included
in
fade
as
a
user
interface
for
the
authentic
similar
with
traditional
machine
learning,
we
can
write
a
training
code
using
passion
in
notebook,
such
as
preparing
data
and
setting
up
the
training
job.
The
algorithm
used
is
called
hospital,
logistic
regression.
It
is
also
called
homo
logistic
regression.
It
is
a
little
complicated
to
improvement.
This
algorithm
faith
provides
a
lot
common,
valid
learning
algorithm,
which
we
can
directly
use
for
our
job
and
made
the
training
as
a
composite
pyrite.
We
run
it
step
by
step
here.
B
A
B
Data
privacy
with
leakage
in
fml,
the
process
of
federated
learning
only
exchanged
encrypted
gradients.
What's
more,
the
change
that
related
model
can
be
converted
into
other
model
formats,
such
as
omnis
by
touch
and
teflon.
Next,
we
can
publish
it
to
our
cave
serving
service
and
serve
it
from
there.
B
B
Can
manage
local
data
contribute
to
the
project?
The
model
we
obtained,
the
collaboration
training
project
a
project
is
a
concept
to
manage
the
correlation
between
different
parties.
It
consists
of
participants,
data
and
jobs.
We
can
manage
different
feature
learning
projects
in
this
unified
interface.
A
job
is
the
fml
task.
We
want
to
calculate
it.
It.