<s>
AlphaGo	B-Application
versus	I-Application
Ke	I-Application
Jie	I-Application
was	O
a	O
three-game	O
Go	O
match	O
between	O
the	O
computer	O
Go	O
program	O
AlphaGo	B-Application
Master	I-Application
and	O
current	O
world	O
No	O
.	O
</s>
<s>
1	O
ranking	O
player	O
Ke	O
Jie	O
,	O
being	O
part	O
of	O
the	O
Future	B-Application
of	I-Application
Go	I-Application
Summit	I-Application
in	O
Wuzhen	O
,	O
China	O
,	O
played	O
on	O
23	O
,	O
25	O
,	O
and	O
27	O
May	O
2017	O
.	O
</s>
<s>
The	O
version	O
of	O
AlphaGo	O
in	O
this	O
match	O
was	O
AlphaGo	B-Application
Master	I-Application
,	O
the	O
one	O
that	O
defeated	O
top	O
pros	O
in	O
60	O
online	O
games	O
,	O
using	O
four	O
TPUs	B-Device
on	O
a	O
single	O
machine	O
with	O
Elo	O
rating	O
4,858	O
.	O
</s>
<s>
DeepMind	B-Application
claimed	O
that	O
this	O
version	O
was	O
3-stone	O
stronger	O
than	O
the	O
version	O
used	O
in	O
AlphaGo	B-Application
v	I-Application
.	I-Application
Lee	I-Application
Sedol	I-Application
.	O
</s>
<s>
AlphaGo	B-Application
Master	I-Application
was	O
actually	O
the	O
second	O
best	O
version	O
that	O
DeepMind	B-Application
had	O
at	O
the	O
time	O
,	O
for	O
it	O
was	O
already	O
in	O
possession	O
of	O
AlphaGo	B-Application
Zero	I-Application
,	O
a	O
version	O
much	O
stronger	O
than	O
the	O
Master	O
version	O
;	O
this	O
can	O
be	O
known	O
by	O
the	O
fact	O
that	O
Nature	O
received	O
their	O
paper	O
on	O
AlphaGo	B-Application
Zero	I-Application
on	O
April	O
7	O
,	O
before	O
the	O
games	O
with	O
Ke	O
Jie	O
.	O
</s>
<s>
DeepMind	B-Application
did	O
not	O
reveal	O
the	O
existence	O
of	O
AlphaGo	B-Application
Zero	I-Application
until	O
the	O
paper	O
was	O
published	O
in	O
Nature	O
in	O
October	O
2017	O
.	O
</s>
<s>
Before	O
the	O
Future	B-Application
of	I-Application
Go	I-Application
Summit	I-Application
,	O
AlphaGo	B-Application
Master	I-Application
defeated	O
Ke	O
Jie	O
by	O
three	O
to	O
zero	O
during	O
its	O
60	O
straight	O
wins	O
in	O
the	O
online	O
games	O
at	O
the	O
end	O
of	O
2016	O
and	O
beginning	O
of	O
2017	O
.	O
</s>
<s>
Google	B-Application
DeepMind	I-Application
offered	O
$1.5	O
million	O
winner	O
prizes	O
for	O
this	O
match	O
while	O
the	O
losing	O
side	O
took	O
$	O
300,000	O
for	O
participating	O
in	O
the	O
three	O
games	O
.	O
</s>
<s>
After	O
the	O
match	O
between	O
AlphaGo	O
and	O
Ke	O
Jie	O
,	O
AlphaGo	O
retired	O
while	O
DeepMind	B-Application
continued	O
AI	O
research	O
in	O
other	O
areas	O
.	O
</s>
