►
From YouTube: # 12: Jeff Hawkins On Defining Intelligence
Description
From the Numenta On Intelligence Podcast series.
B
C
C
B
B
B
A
E
A
C
A
C
B
C
B
B
Basically,
for
so
years
and
and
then
when
things
started
to
move
movement
requires
there's
two
ways
you
can
move,
you
can
move
blindly.
You
know
it's
just
pretty
stupid,
no
point
moving
and
we
don't
really
you
know,
but
if
you're
gonna
move
with
a
purpose
to
achieve
something,
you
have
to
have
some
sort
of
knowledge
about
the
world
and
that
knowledge
could
be
encoded
in
genes,
the
genetic
material,
but
it's
some
sort
of
mouth,
even
a
bacterium.
B
It
has
a
motor
plan,
which
is
you
know,
I
can
move
and
I'll
move
I'll
turn
towards
gradients
that
you
know.
That's
my
that's
my
strategy,
it's
my
model
of
the
world,
but
any
any
animals
that
start
moving
right.
A
lot
then
they
need
they
need
to
have
some
sort
of
concept
of
where
they
are
and
where
things
are,
and
so
all
of
a
sudden.
B
If
you're
going
to
have
mobility,
then
you're
gonna
have
to
have
some
sort
of
way
of
learning
the
structure
of
your
environment
and
know
where
you've
been
how
to
get
back
to
some
place
that
begun
on
a
home
any
other
way
from
any
kind
of
come
back.
You
have
to
have
some
way
of
doing
that
right
and
so
nature
evolved.
A
series
of
ways
of
sort
of
building
maps
of
your
environments
and
different
animals
do
the
different
ways,
right
and
in
mammals.
B
A
B
B
Well,
yeah
I,
don't
want
to
use
those
terms,
be
it's
navigating
into
some
rooms.
I
have
to
learn
like
where's
where's,
my
environment,
with
whether
it's
a
woods
or
my
house,
or
the
rats
being
left
on
your
house
or
whatever.
It
is
right,
learning
that
where
things
are
and
how
to
get
around
and
navigating
whether
it's
and
and
so
there's
it's
complex.
That
system
is
very
complex
to
do
this
right.
B
There's
egocentric
and
allocentric
models
there
yeah
in
the
in
Toronto
cortex,
so
that,
but
again,
that's
been
on
the
evolutionary
pressure
for
a
very
long
period
of
time.
Of
course,
in
the
new
your
cortex,
which
we
think
of
as
this
is
the
organ
of
intelligence,
is
fairly
new.
It
hasn't
got
big,
really
rapidly
yeah,
and
so
the
idea
they
got
big,
really
rappin
it
by
taking
a
single
thing
and
making
many
copies
of
it.
A
B
A
B
B
So
that
was
my
Auburn
and
mountain
council
is
a
big
proposal
in
the
1970s.
Was
that
the
you
know
the
New
York
cortex
got
big
rapidly
and
and
but
you
look
at
the
structure,
the
structure
is
very
similar
everywhere,
so
he
says
this
is
just
it's
the
same.
Algorithm
being
you
know,
operated
on
a
hundred
thousand
times.
No,
we
have
a
hundred
thousand
two
hundred
thousand
copies
of
the
same
circuit,
but.
B
This
is
a
thousand
brains
theory,
which
of
course,
is
in
the
December
paper
yeah.
It
doesn't
make
thousand
brains,
New,
York
intelligence
says
you
got
all
these
parallel
months,
so
basically
doing
is
tweaks
on,
but
they're
basically
doing
the
same
thing.
But
what
thing
they
can
model
different
things
they're,
not
all
my
there
are
all
these
models
are
different.
B
It
doesn't
know
what
these
combos
in
the
word
is
doing
great,
but
depending
on
what
its
inputs
are
and
what
motor
behaviors
it
could
control
that's
what
it
builds
a
model
and
some
of
them
are
going
to
be
building
models
of
input
from
your
eyes
and
your
fingers.
Some
are
gonna,
be
building
models
up
already
the
mock-up
models,
models
and
models
that
have
been
getting
input
from
other
parts
of
New
York.
What
Jack's
right.
D
B
A
B
Yeah
so
I
think
this
is
one
of
the
I
think
today
we're
in
this
sort
of
a
weird
state
about
intelligence,
especially
machine
intelligence,
which
is
dominated
by
the
idea
that
intelligence
is
a
capability,
can
I
do
something.
Can
I
play
go
better
than
the
best
human
player?
Can
I
Drive
a
car
better
than
you
know,
I'm
a
human
or
can
I?
You
know
whatever
pick
your
favorite
task,
you
know
analyze.
B
B
Just
a
quote,
you
know
against
human
level.
Behavior
right
take
a
task
that
a
human
does
right.
Take,
can
I
get
a
machine
to
do
that
task,
and
this
goes
back
to
Turing
with
the
Turing
test
right.
He
he
proposed
this.
The
imitation
game,
which
we
now
call
the
Turing
test
right,
but
which
is
all
about
matching
human
level
performance
right,
but
I'd
only
has
acts
of
really
poor
way
of
measuring
intelligence.
B
B
B
B
C
B
B
B
It
the
point,
is
you
know
you?
If
you
have
some
input
and
you're
going
to
act
on
it,
then
you
need
a
model
to
decide
how
to
act
right
being
really
stupid
me
a
lookup
table.
It
could
be,
but
it's
a
here's
an
instructions
on
how
to
do
this
right
or
it
could
be
learned.
But
the
point
is
our
brains.
We
have
this
fairly
general-purpose
way
of
learning
miles
the
world
right.
My
brain
can
learn
mathematics.
It
can
learn
Portuguese
if
I
so
desire
to
do
so.
I
come
in
to
play.
B
C
B
Donnelly
using
the
same
basic
algorithm
and
now
we've
got
we
do
them
all
simultaneously.
It's
like
it's,
not
I'm,
not
dedicated
than
one
thing
so
I
have
a
I.
Have
this
you've
got
a
picture
in
your
head
in
your
head
right
now,
you've
got
this
model
of
the
world.
I
know
you're
interested
in
this.
You
know
still
astronomy
and
where
it's
didn't
science.
B
Space
and
and
then
you
also
have
a
model
of
your
family
in
your
car
and
have
fixed
the
things
in
your
house
and
all
these
things.
So
we
have
this
very,
very
rich
model
that
we've
learned
and
a
single
organ.
Then
your
cortex
learns
can
learn
this
very,
very
rich
model
of
the
world.
Yes,
if
we
want
to
ask
how
intelligent
something
is
we
we
really
need
to
ask
what
is
its
mother
like
well
and
how?
Which
is,
and
by
what
methods?
Does
it
work?
B
C
B
I
set
up
a
go
playing,
adjusted
chest
if
I
set
up
a
chess-playing
computer
I
want
to
build
the
world's
best
chess
player
better.
The
model
may
be
built
around
chess.
It
may
be
like
oh
there's,
the
only
rule
is
this
thing
can
make
our
chess
board
movements
and
the
only
thing
you
know
our
chess
board
positions,
and
so
we
we
structure
them.
The
information
in
this
computer
in
chess
board
coordinates
right.
B
That's
right,
it
said
we
talk
about.
You
know
good
cells
like
a
like
a
frame,
a
reference
frame
right
and
but
it's
a
very
general
purpose.
Reference
map,
like
imply
grid
cells.
All
these
different
things,
it's
more
like
XYZ,
Cartesian,
coordinates.
It's
very
kind
of
purpose.
You
can
apply.
Xyz
coordinates
to
lots
of
different
things,
so.
A
B
B
Yeah,
you
need
to
know
where
things
are
relationship,
the
other
things
and
that's
what
you
need
to
know.
Basically,
that's
how
you
assemble
information
to
something
useful
like
it.
You
know
what
makes
a
computer
or
computer
is
because
these
components
that
are
relation
to
each
other
and
how
they
move
relative
to
each
other
and
so
on
right.
So
you
need
a
reference
frame
for
storing
structure
about
anything
and
there's
different
types
of
reference
names.
You
know
a
good
general
purpose.
Reference
frame
isn't
one
I
just
mentioned
the
XYZ
one
parties
et.
B
A
that's
pretty
general
purposes.
You
can
apply
to
any
three
dimensional
or
two
dimensional
one
dimensional
structure
and
you
can
add
more
dimensions
if
you
want
yeah.
Grid
cells
are
another
general
purpose
model
there's
in
that
regard,
but
they
work
differently
than
the
coordinates
in
clever
way
and
just
a
little
aside
here.
The
reason
they're
really
clever
is
because
there
there
is
no
origin
to
them,
but
the
locations
are
tied
together
by
movement,
and
so
it's
all
it's
a
it's
like
a
it's
like
a
Cartesian
coordinate
frame
in
some
sense
and
that
it's
your
purpose.
B
But
you
can
think
of
it
as
a
general-purpose
reference
frame,
so
I
can
I
can
use
a
general
perfect
reference
frame
to
learn
chess
I
can
say:
oh
well,
there's
a
set
of
a
board
here.
Maybe
I'll
use
my
set
of
X
like
confusion
coordinates
for
that
sure,
and
it
won't
be
as
good
as
a
chess
coordinate
frame
yeah.
B
B
There's
no
encode,
there's
no
assigned
reference
frames,
there's
no
even
knowledge.
It's
essentially
has
to
be
learned,
and
but
it's
no
one
really
understands
how
it's
learned
and
so
II.
You
end
up
with
this
sort
of
weird
sort
of
mapping
between
inputs
and
outputs
that
no
one
really
understands
right.
Well
right.
So
the
fact
that
there
is
a
mapping
between
inputs,
outputs
tells
me
there
has
to
be
some
sort
of
model
whether
that
model
has
a
reference
frame.
B
B
C
B
A
Right
and
you're
talking
about
so
there's
different
types
of
reference
frames,
yeah
like
I
like
to
think
there
isn't
there's
the
easy
separation
of
I
have
an
egocentric
reference
frame.
I
am
in
the
center
of
it,
and
my
environment
is
the
frame
of
mark
where
my
organism
exists
and
then
there's
the
reference
frame
of
every
object.
That
I
could
imagine
that
has
its
own
reference
frame,
yeah
and
then
I
can
somehow
combine
them.
So
I
can
imagine
an
object
in
my
hand.
That's
not
there.
Let.
B
B
A
B
A
B
C
D
B
What
we
think
is
going
on
in
the
way
of
regions
in
the
brain
right,
then
you
can
have
reference
frames
that
are
anchored
to
physical
objects,
meaning
if
the
object
moves,
the
reference
frame
moves
with
it
right,
that's
what
we
think
is
going
on
in
what
regions
and
they
are
cortex.
Then
there
are
yeah.
B
So
you
know,
there's
excuse
me,
there's
these
well-known
when
people
first
discovered
the
what
and
where
regions
in
the
neocortex.
Oh
yeah
these
these
parallel
sensory,
perhaps
for
vision,
hearing
and
touch,
and
they
first
discovered
a
vision,
and
it
was
very
odd
because
if
some
of
the
human
has
a
damaged
what
halfway
in
vision,
yeah.
C
B
B
B
So
they're
like
oh,
it
was
a
coffee
cup,
of
course,
why
you
know,
but
the
point
is
with
it:
with
a
with
only
the
ego
centric
reference
frame.
You
can
know
how
to
reach
things,
and
you
can
know
where
things
are
relative.
You
cannot
recognize
what
this
thing
is
because
of
it.
You
need
me
an
object
since
you've
reference
images
you
flip
around
and
if
people
would
have
a
damaged,
where
pathway
but
a
functioning,
what
pathway
they
say.
B
B
So
when
we
loose
a,
we
model
the
world
it
to
be
careful
here,
I'm
on
with
a
coffee
cup,
with
ego
set
up
with
the
object,
centric
reference
frames,
sure
and
I
I
modeled
the
solar
system
that
way
and
I
modeled
the
you
know
the
universe
I'll.
You
know,
because
I
can't
actually
go
out
and
touch
those
things
yeah,
but
but
to
be
a
functioning
animal
and
with
moving.
B
Have
to
have
egocentric
reference
frames,
yeah,
and-
and
so
you
know
so
you
know
I.
Would
you
take
a
primitive
animal
I
always
like
to
talk
about
like
a
crocodile,
and
you
know
they
see
some
food.
They
may
not
have
a
clear
image
of
what
that
food
is
or,
however,
you
know
exactly
its
shape
and
width,
you
know,
but
they
know
how
to
reach
it.
Yeah.
C
B
Move
the
face
towards
it
and
you
know,
bite
the
thing,
so
it's
really
important
to
be
able
to
like
move
your
body
to
capture
some
prey
or
something
like
that.
So
there's
these
limited
abilities
to
do
these
things
and
other
animals.
So
these
these
ideas
aren't
just
in
neocortex
they're
in
the
old
parts
of
the
brain
too.
A
B
C
B
B
B
B
Of
course,
but
the
idea
that
they
can
see
the
structure
that
object,
what
that
tells
you
is
that
they
see
the
structure
of
the
thing
they're
about
to
eat.
They
can,
then
they
know
they
recognize
its
orientation
and
what
features
were
on
it
and
then
they
can
move
their
hand.
They
grab
it
in
the
right
way
to
bring
it
to
their
mouth
in
the
right
way
to
eat
it,
and
so
it's
not
like
it's
just
some.
B
B
I,
don't
know
there's
so
many
things
are
confused
here.
So,
first
of
all,
we
can
ask
the
mechanisms
by
which
the
animal
makes
a
model
of
the
world
are
those
general-purpose
mechanisms
or
they
specifically
then.
E
B
Ask
what
is
the
capacity
of
that
model,
because
clearly
a
mouse's
neocortex
is
quite
small
and
yes,
the
size
of
a
small
postage
stamp
at
best,
where
rats
about
that
big,
so
nice
is
probably
even
smaller,
so
it
doesn't.
It's
not
gonna
have
a
very
big
model
of
the
world,
but
it
learns
that
model
the
same
way.
You
and
I
learn
about
a
general
purpose
model,
but
very
limited
capacity
right,
whereas
you
know
innate,
it's
got
a
much
bigger
model
like
you
know
a
monkey
and
or
and
then
use
humans.
B
We
have
this
really
big
airport
XCOM,
so
we
all
learn
the
same
way.
We
all
use
the
same
mechanism,
we're
all
general-purpose
learners.
We
all
learn.
You
know
we
learn
rapidly
continuously
and
using
the
sensory
motor
inference
and
we
have
the
same
sort
of
rich
modeling
structure
right,
but
clearly
so
we
have
to,
we
can
make.
We
can
divide
animals
in
law
or
systems
along
that
line.
What
is
the
mechanism
you're
using
to
learn
and
then
is
a
capacity
issue?
Yeah.
So
often
we
think
about
you
know.
Intelligence
is
awesome.
B
People
who
know
a
lot
of
things
right.
Well,
that's
a
separate
issue:
that's
like
okay!
Well,
a
mouse
I
would
say,
has
the
same
sort
of
mechanisms
of
learning
that
we
do
it's
a
mammal,
and
so
it's
going
to
exact
same
neocortical.
Structures
are
very
similar,
but
it's
just
limited.
Sometimes
I
make
the
analogy
with
computers
right
there
is.
What
is
a
computer
well.
Is
that
there's
a
formal
definition
for
a
universal
Turing
machine?
Sure
it's
a
mathematical
thing.
B
E
B
And
and
all
computers
that
we
think
of
today,
as
computers
are
universal
Turing
machines,
not
all
you
can
build
an
ASIC
chip
that
does
something
it's
not
even
emotional
Turing
machine,
but
computers
are
yeah,
but
the
computers
come
in
all
these
different
sizes
right,
you
can
get
the
tedious
little
computer,
still
a
pronestyl
little
cpu.
So
this
program,
and
maybe
only
eight
bits
right-
has
a
limit
amount
of
memory.
Hours,
you're.
E
B
C
B
So
it
did
sort
of
the
same
functions,
but
it
wasn't
as
good,
but
there's
no
computer
inside
of
it.
Thank
you
so,
but
it
didn't
many
of
the
same
things.
So
so
then
II
way.
So
you
have
this
one
spectrum,
which
is
the
mechanisms
that
are
being
used
and
in
the
computer
world,
that's
universitary
machines
and
then
there's
assassin
section
of
capacity
and
then
there's
a
third
thing
which
is
like
okay.
What
is
it
been
programmed
to
do
so
now
we
can
have
this.
We
have
similar
three
similar
metrics
on
intelligence.
B
We
have
a
general-purpose
reference
frame
and
we
and
we
the
whole
mechanism
that
we
use
in
building
a
model
the
world.
This
is
what
we
wrote
in
the
paper
yeah,
and
so
we
all
be.
We
dis
all
use
the
same.
One
right,
then
there's
a
capacity
issue,
so
Mouse
is
very
limited
capacity.
We
have
much
more
capacity,
cuz,
it's
just
bigger
brains
and.
B
A
B
Well,
whether
it's
an
ease
when
we
think
about
intelligence,
we
have
to
think
along
these
metrics.
That's
the
same
thing
with
computers
with
you
know
those
three
metrics
for
computers,
the
method
in
which
it
operates
the
capacity
of
the
system,
and
then
what
has
been
programmed
to
do,
and
it
tells
us
we
have
the
method
of
learning
a
model
which
is
reference
frames
and
general
purpose
and
sensory,
motor
and
fist
all
those
things
then,
because
they
can
vary
there
yeah.
So
that
may
be
the
goal.
Computer
doesn't
have
a
general
purpose
reference
frame
right.
B
Then
there
is
the
capacity
of
the
system.
How
much
memory
does
it
have
have
bigger
brains
or
how
many
columns
do
I
have
I'm
eating
neurons?
How
many
synapses
things
like
that,
and
then
there
is
what
is
it?
What
is
it
been
trained
to
do?
What
does
it
learn
right,
and
so
you
and
I
could
be
very
intelligent.
People
have
very
different
sets
of
knowledge
about
the
world,
so
if
I
had
never
been,
if
I
was
raised
in
the
woods
by,
you
know,
wolves
and
you
came
to
on
with
Jeff.
B
B
A
B
Three
components:
method
by
which
it
works:
mm-hmm.
Is
it
general-purpose
or
not?
Have
you?
Yes,
we
have
a
general-purpose
learning
algorithm
for
learning
models,
the
world
or
not,
then
you're
passing
the
system
and
then
what
it's
been
trained
and
how
it
learns
right
and
so
right
now
we're
focusing
on
hey.
Can
I
beat
some
humans
at
something
right,
which
is
really
the
wrong
metric
completely
yeah.
C
B
C
B
A
Along
those
lines,
I
want
to
talk.
We
just
talked
about
the
intelligence
spectrum.
I
would
assume
that
we
humans
are
probably
near
the
top
of
that.
Tell.
B
A
B
C
B
Your
cryptic
looks
the
same
everywhere.
Therefore,
these
things
are
based
on
reference
range.
It's
like,
let's
figure
out
how
right,
as
opposed
to
Sargent,
saying
you
know,
I
think
language
is
based
on
not
reject.
Now
I
was
like
damn
it.
Language
must
be
built
on
reference
frames.
So,
let's
think
about
it,
a
bit
yeah.
B
B
D
B
B
B
Knowledge,
let's
just
I'm,
just
sort
of
the
final
acknowledges
knowledge
is
sort
of
information
or
facts
arranged
in
a
useful
way.
Correct.
Okay,
so
I
can
have
a
whole
bunch
of
facts
and
I'll.
Tell
ya
you
and
I
see
the
same
facts,
but
if
I'm
not
about
it,
I.
Look
at
those
facts.
Oh
I
know
those
things
and
what
does
it
mean
to
know
those
things,
but
what
I
believe
it
means
to
know
those
things
is
that
I
take
that
information
and
I
can
assign
it
to
a
reference
right.
B
Everything
has
a
location
that
reference
frame
and
when
you
get
with
a
reference
frame,
is
you
get
the
ability
to
navigate
through
these
facts?
Then
you
navigate
through
the
facts
is
like
navigating
through
a
room
I'm
moving
my
finger
off
with
a
coffee
cup,
but
in
this
case
I'm
moving
my
location
in
this
space
of
facts
or
space
of
things
to
achieve
certain
results.
Its
own
example
I've
often
used
as
a
mathematician
where
I
say.
Okay,
some
people
look
at
mathematics.
B
B
Once
they
sign
that
searching
cube,
oh
yeah,
you
know
so
and
what
they
do
is
so
the
one
thing
you
think
about
is
like,
if
I'm
trying
to
do
trying
to
solve
a
mathematical
theorem
or
something
like
that,
I
start
with
set
of
points.
The
set
of
equations
or
a
set
of
things.
I
know
true
right
that
even
some
reference
frame
and
I'm
trying
to
get
to
a
new
point
in
some
space
and
I'm
trying
to
figure
out
the
right
behaviors
to
get
me
there
and
the
behaviors
of
mathematical
operations.
B
A
it's
a
it's
literally
using
grid
cells
yeah,
but
the
space
it's
just
two
things:
what
are
the
think?
It
was
good
so
like
a
map,
okay,
right,
n,
dimensional
map.
So
what
are
you
assigning
to
the
map
locations?
You're?
Signing
mathematical
constructs.
You
can
think
about
good
questions:
okay,
okay,
yeah,
when
you
move,
what
are
you
moving?
We're
not
moving
physically,
but
you
are
mentally
moving
through
this
space
by
applying
certain
operators
instead
of
bringing
me
instead,
the
operating
being
flexed.
This
muscle,
the
operator
is
multiplying
yeah.
B
And
you
get
the
same
basic
thing:
you're
moving
through
the
space
of
ideas,
and-
and
so
that's
what
you
do
when
you're
thinking
about
something
so
this.
So
this
is
the
general
idea
that
high-level
knowledge
is
the
organization
of
concepts
in
a
reference
frame
and
and
your
ability
to
be
really
really
smart
at
something
is
knowing
how
to
navigate
and
what
kind
of
behaviors
you
do
to
get
someplace.
Another
example
would
be
a
politician.
B
C
B
B
And
they
have
all
these
potential
things
they
could
do.
They
could
have
a
rally,
they
could
hold
the
forum,
they
could
do
some
sort
of
publicity,
they
could
get
some
endorsements
and
they
and
the
politicians
experienced
enough
to
say
they
know
what
will
happen
when
they
do
these
different
behaviors
right
right
in
so
the
sense,
your
choice
of
behaviors
is
your
choice
of
which
way
you
want
to
move
through
the
space
of
political
things.
B
B
That
field
yeah
right
more
likely
a
woman
because
they're
smart
about
these
things,
yeah
what
you
gave.
Would
you?
What
becomes
an
expert
is
an
expert.
You
can
look
at
the
same
facts,
but
they
have
through
experience,
they've
learned
how
what
happens
when
you
do
things
right.
So
if,
as
a
programmer,
you
sit
there
and
you
say:
I'll
come
on
solve
some
problem.
Yeah.
C
B
B
B
Of
know,
you
have
a
reference
frame,
it's
just
like
seeing
when
you
see
this
coffee
cup,
you
don't
say
where's
my
reference
saying
you
said:
that's
a
coffee
cup.
Well,
a
person
who
spent
a
lot
of
time
in
the
woods
will
look
at
that
same
set
of
day-to-day.
You
look
at
it
right
and
understand
it
right
and
you'll
say
oh
yeah
well,
I
know
how
to
do
that.
I
know
how
to
get
to
the
water,
because
water
is
typically
getting
to
me.
B
This
kind
of
thing
down
there
right
yeah
and
you
because
you
know
that
right
and
and
same
thing
if
you've
never
been
exposed
to
something.
Then
everyday
object,
like
a
coffee
company
raised
in
the
woods,
never
saw
any
kind
of
physical
thing
like
this.
You
would
might
look
at
ago
I
kind
of
see
what
it
is,
but
you
wouldn't
immediately
go.
That's
gonna
be
great
for
carrying
the
the
fuel.
We
need
the
car
you
know
or
whatever
way.
B
B
Know
different
people
could
take
the
same
set
of
facts.
It
organized
them
in
a
different
reference
names
absolutely,
and
we
see
this
a
lot
yeah
that
doesn't
happen
for
everyday
objects
as
much,
but
we
see
it
for
things
that
are
conceptual,
and
so
we
can,
you
know
different
religious
beliefs
take
the
same
facts
about
the
world.
We
all
observe
the
same
thing
and
they
come
to
completely
different
conclusions
about
what
would
happen
when
they
do
certain
things.
Right
and-
and
that's
that's
a
case
where
you
know
the
reference
frames,
aren't
reality
right.
B
They
can't
all
be
correct.
Mostly
you
know
it's
all
the
different
beliefs
we
have
in
the
world
can't
all
be
they
conflict
this
one.
No,
they
cannot
all
be
correct,
so
most
one
can
be
dragged
probably
most
of
it
wrong.
Oh,
but
but
the
point
is
you
know,
that's
that's
the
test
of
reality
is
or
testament
the
accuracy
of
a
model
is
how
well
it
predicts
the
future
and
and
when
we
don't
really
have
good
data
about
the
future.
C
C
C
C
B
B
Yours,
beliefs,
it's
your
reality.
This
is
what
you
know.
That's
it!
That's
what
it
boils
down
to.
There
is
no
more
the
reality
and
in
terms
of
your
head,
this
is
it
yeah
instead
of
you
say.
Well,
you
know
what
everything
you
believed
it's
these
references
and
you
have
to
start
over
again
then
it's
kind
of
like
oh
you're,
going
back
to
square
one
in.
A
B
B
B
B
C
B
C
B
B
Now
it's
like
it's
like
feeling
lost,
you
know
if
you,
if
you're,
if
you're
out
in
the
woods
and
all
of
a
sudden
there's
you
can't
anchor
yourself
in
the
woods
every
tree
looks
the
same,
and
now
your
reference
frame
is
lost
you
little.
If
you
cannot
locate
yourself
in
the
reference
frame,
you
feel
lost
absolutely.
B
E
C
C
B
This
looks
somewhat
similar,
that's
right!
So
we
have
these
reference
frames.
You
I
believe
models
the
world.
We
want
to
stick
new
data
into
our
models
of
the
world
and
by
and
by
if
you
have
to
abandon
one
of
your
basic
models
of
how
the
reality
is,
then
you
feel
lost
and
that's
an
uncomfortable
feeling
and
you
don't
know
what
to
do
when
you
and
it's
like
being
lost
in
the
woods
so.
A
B
B
B
B
C
B
B
B
A
B
Fine,
but
you
know
but
I
mean
but
I
can't
take.
You
know
the
world's
largest
weather
simulation
and
run
it
on
the
computer
in
my
toaster
or
if
I
can't
do
it
in
any
kind
of
real
time,
that'd
be
useful
right.
So
so
those
are
sort
of
the
three
metrics
and
we
just
confuse
them
all
the
time,
yeah
and
and
I
think
it's
even
worse
than
that.
We
confuse
them
cuz.
We
only
look
at
most
AI
looks
at
what
humans
do
right.
B
Only
what
humans
do
and
then
even
then,
like
the
Turing
test,
brings
us
to
a
whole
nother
domain,
which
is
like
well
now
you're,
trying
to
emulate
the
emotional
capabilities
of
a
human
and
my
definition
of
intelligence.
We
just
talked
about
does
not
include,
emotions
doesn't
mean
human-like
right,
it
doesn't
mean
you
know
it's
more
spot
like
then,
then
you
know
you
know,
did
I
get
that
right.
Yes,.
D
C
B
A
C
B
C
B
B
B
I,
basically,
what
is
the
structure
of
the
world
and
can
I
discover
that
structure
of
the
world,
whether
it's
physical
objects
in
or
mathematics,
or
you
know
politics
whatever
I'm
gonna,
try
to
learn
the
structure
of
the
world
in
an
actionable
form
so
that
I
can
now
see
new
things
place
them
in
this
model
know
how
to
act
related
to
them
and
so
on.
But
those
are
the
three
basic
things
and
we
don't.
B
B
How
did
it
learn
it
and
and
all
those
thing,
and
until
we
not
ask
you
these
questions,
no
one's
sitting
around
saying
what
people
do
point
out
that
the
the
chest
plane
computer
doesn't
seem
to
know
anything
else
yeah,
but
they
don't
ask
you
don't
have
that.
Why?
Why
didn't
know
anything
else,
because
it's
got
a
chest
reference
frame
and
that's
all
that's
ever
been
trained
on
and
they
probably
can't
learn
anything
else
anyway.
That's
it's
not
a
simple
one.
Binary
answer,
there's
those
three
things
all
have
their
separate
dimensions.