<s>
Multiple-try	B-Algorithm
Metropolis	I-Algorithm
(	O
MTM	O
)	O
is	O
a	O
sampling	O
method	O
that	O
is	O
a	O
modified	O
form	O
of	O
the	O
Metropolis	B-Algorithm
–	I-Algorithm
Hastings	I-Algorithm
method	O
,	O
first	O
presented	O
by	O
Liu	O
,	O
Liang	O
,	O
and	O
Wong	O
in	O
2000	O
.	O
</s>
<s>
In	O
Markov	B-General_Concept
chain	I-General_Concept
Monte	I-General_Concept
Carlo	I-General_Concept
,	O
the	O
Metropolis	B-Algorithm
–	I-Algorithm
Hastings	I-Algorithm
algorithm	I-Algorithm
(	O
MH	O
)	O
can	O
be	O
used	O
to	O
sample	O
from	O
a	O
probability	O
distribution	O
which	O
is	O
difficult	O
to	O
sample	O
from	O
directly	O
.	O
</s>
<s>
Multiple-try	B-Algorithm
Metropolis	I-Algorithm
needs	O
to	O
compute	O
the	O
energy	O
of	O
other	O
states	O
at	O
every	O
step	O
.	O
</s>
<s>
It	O
can	O
be	O
argued	O
that	O
this	O
method	O
only	O
appears	O
faster	O
because	O
it	O
puts	O
much	O
more	O
computation	O
into	O
a	O
"	O
single	O
step	O
"	O
than	O
Metropolis-Hastings	B-Algorithm
does	O
.	O
</s>
