►
From YouTube: Numenta research paper review: Layer 6 ensembles can selectively regulate the behavioral impact...
Description
Review of paper in bioarxiv: https://www.biorxiv.org/content/10.1101/657114v1
Layer 6 ensembles can selectively regulate the behavioral impact and layer-specific representation of sensory deviants
Broadcasted live on Twitch -- Watch live at https://www.twitch.tv/rhyolight_
B
B
A
B
B
B
This
is
not
an
easy
paper
to
read:
there's
a
lot
of
details
in
it.
They
do
summarize
it,
but
it's
still
difficult.
That's
a
constant
summary,
so
they
just
give
you
the
big
picture
here.
Big
picture,
the
good
thing
about
it
is
they're
trying
to
understand
effectively
our
sixth
on
other
layers
in
predictive
model.
It
does
sounds
pretty
close
to
our
goal.
B
It's
not
as
close,
do
you
think,
partly
because
they're
really
restricted
in
many
ways
than
they,
but
they
can
do
so.
It's
a
vicious
but
they're
doing,
but
it's
also
pretty
far
from
I
guess
so,
for
example,
in
the
predictive
model,
they
start
right
at
the
very
beginning,
here
very
first
line,
the
abstract
enhances
the
sailing
team
I'm,
anticipating,
but
seldom
you
notice
the.
B
B
We
can
change
this
legends
while
the
rats
feeling
it
and
they
want
to
make
sure
the
rock
can
hear
that
I
want
to
feel
the
air
movement,
so
they
saw
it
assuming
it's
in
the
dark,
because
otherwise
I
could
see
it
well,
I
don't
know
anyway,
the
result
is
the
following,
so
they
have
the
strong
signal
and
weak
signal
again.
Optogenetic
modifying
photo
six
and,
for
example,
in
this
particular
case
they
say.
B
B
C
B
B
B
C
B
B
A
C
B
B
This
case,
just
trying
to
psych
me
get
across
you
want
to
get
over
there
and
first
thing
we
do
is
say:
hey
well,
I'm
simply
just
strongly
activate
this,
and
then
you
get
part
of
them
to
show
that
when
they
leave
the
activated,
it
doesn't
happen.
Okay,
so
the
same
there's
a
threshold
above
which
I
can
make
right
before
hooli
and
now
we're
talking
other
below.
B
B
B
B
B
B
B
And
right
now
it
is
a
deviance
detection,
not
have
changed,
call
it
change,
Batali,
there's
a
static
and
put
on
some
sensor,
and
then
it
changes,
there's
a
dynamic
input
in
the
sensor
and
then
their
nature.
The
dynamic
input
changes
I
mean
they're.
Throughout
this
paper
there
are
so
many
little
caveats
in
every
single
result.
You
just
really
got
it.
You
know
it's
just
every
little
one
like
so,
though,
there's
being
boxed
in
all
over
the
place
here
sweet.
B
This
is
the
introduction
of
the
second
experimental
paradigm.
We
do
politically
chelation
Iran
change,
identity,
endless
different
ensembles,
basically
saying
get
himself
back
to
me.
Without
you
to
mean
my
final,
a
cosplayer
so
basically
mean
fine,
makes
all
the
other
layers
activated
later
six,
it
changed,
but
it
was-
and
this
is
they
disobey
determined.
By
doing
this,
tetra
recording
and
their
ego
spiking
their
honor.
The
jour
is
rendering
fast
bike.
You
know,
journalist
in
problems.
B
B
B
A
B
B
B
B
B
C
B
C
B
B
B
B
B
They
made
a
point
here
in
this
paper
somewhere.
They
make
the
point
here
that
these
exposures
have
been
done
primarily
in
the
past,
with
like
audio
and
if
you
use
all
the
other
there's
a
lot
of
pre-processing
done
on.
So
some
of
this
deviance
detection
could
have
been
pre
pre
cortical
humans,
being
s1
is
more
direct,
so
they're
saying
I
was
going
to.
B
In
contrast
here
them
contrast,
heavy
feeders
and
coating
and
cooking
still
in
US
history,
emergent
literacy,
even
on
small
ex-muslim
example,
to
get
your
curly
positive
farmers
in
their
formative.
Six,
the
TBM
increases
during
during
high
poverty
contrast
Heather,
Knutson
coding,
so
the
layer
to
treat
cells
shows
all
the
positive
negative
correlations
with
so
happy.
So
I
have
a
reduced
signal.
Strength
later.
If
they
compete
would
say,
hi
some
cells
are
quite
more
actively
than
and
if
it
increased
when
other
player
B's
of
the
more
actively,
but
that
I
wasn't
sure
my
horn
they're
six.
B
B
B
B
And
then
they
said
this
choices
to
happen.
This
specific
population,
elation
excels,
active
during
a
phonemic
Drive,
was
different
from
lack
of
during
control
conditions,
so
this
changing
relationships
could
lead
to
disruption
of
genders
and
later
to
three,
because
this
is
how
we
would
say,
because.
B
So
I
really
I'm
not
gonna,
go
into
when
he
killed
I.
Just
there's
lots
of
good
love
this,
where
you
just
spent
a
half
an
hour
an
hour
24
the
data,
but
for
me
personally,
when
I
need,
is
unwitting
for
some
insights
and
just
constantly
on
the
radar
like
something
jumped
out
that
I
didn't
find
that
too
much
here.
There's
a
lot
of
things
you
didn't
know,
but
they
didn't
seem
to
really
clearly
relate
to
our
work.
When
I
could
say
you
look
at
the
references
they
had
for
the
circuitry
that
this
was.