<s>
In	O
artificial	B-Application
intelligence	I-Application
,	O
eager	B-General_Concept
learning	I-General_Concept
is	O
a	O
learning	O
method	O
in	O
which	O
the	O
system	O
tries	O
to	O
construct	O
a	O
general	O
,	O
input-independent	O
target	O
function	O
during	O
training	O
of	O
the	O
system	O
,	O
as	O
opposed	O
to	O
lazy	B-General_Concept
learning	I-General_Concept
,	O
where	O
generalization	O
beyond	O
the	O
training	O
data	O
is	O
delayed	O
until	O
a	O
query	O
is	O
made	O
to	O
the	O
system	O
.	O
</s>
<s>
The	O
main	O
advantage	O
gained	O
in	O
employing	O
an	O
eager	B-General_Concept
learning	I-General_Concept
method	O
,	O
such	O
as	O
an	O
artificial	B-Architecture
neural	I-Architecture
network	I-Architecture
,	O
is	O
that	O
the	O
target	O
function	O
will	O
be	O
approximated	O
globally	O
during	O
training	O
,	O
thus	O
requiring	O
much	O
less	O
space	O
than	O
using	O
a	O
lazy	B-General_Concept
learning	I-General_Concept
system	O
.	O
</s>
<s>
Eager	B-General_Concept
learning	I-General_Concept
systems	O
also	O
deal	O
much	O
better	O
with	O
noise	O
in	O
the	O
training	O
data	O
.	O
</s>
<s>
Eager	B-General_Concept
learning	I-General_Concept
is	O
an	O
example	O
of	O
offline	B-General_Concept
learning	I-General_Concept
,	O
in	O
which	O
post-training	O
queries	O
to	O
the	O
system	O
have	O
no	O
effect	O
on	O
the	O
system	O
itself	O
,	O
and	O
thus	O
the	O
same	O
query	O
to	O
the	O
system	O
will	O
always	O
produce	O
the	O
same	O
result	O
.	O
</s>
<s>
The	O
main	O
disadvantage	O
with	O
eager	B-General_Concept
learning	I-General_Concept
is	O
that	O
it	O
is	O
generally	O
unable	O
to	O
provide	O
good	O
local	O
approximations	O
in	O
the	O
target	O
function	O
.	O
</s>
