Re: [Computer-go] Would a larger board (25x25) dramatically reduce AlphaGos skill?

[Sighris]

See below:

On Tue, Mar 22, 2016 at 5:21 PM, Lukas van de Wiel <
lukas.drinkt.t...@gmail.com> wrote:

> It would reduce Alphago, because there is less training material in the
> form of high-dan-games, to train the policy network.
>


Maybe not a concern. There has been a suggestion that AlphaGo be allowed to
learn the 19x19 game with no human games input.
 - Also, there are larger game records (23x23, 25x25, and bigger) on KGS,
which is where the AlphaGo team got the SGF files from to train for Lee
Sedol.


>
> It would also reduce the skill of a human opponent, because (s)he would
> have less experience on a larger board, just as AlphaGo.
>

Agreed Joseki would be different, etc... but I agree with Ray Tayek
that humans can adopt quickly.


>
> It would be fun to see which can adapt better.
>

Agreed 100%!!!
Sighris



>
> Cheers
> Lukas
>
> On Wed, Mar 23, 2016 at 1:18 PM, Ray Tayek  wrote:
>
>> On 3/22/2016 11:25 AM, Tom M wrote:
>>
>>> I suspect that even with a similarly large training sample for
>>> initialization that AlphaGo would suffer a major reduction in apparent
>>> skill level.
>>>
>>

> i think a human would also.
>
>

>
>>The CNN would require many more layers of convolution;
>>> the valuation of positions would be much more uncertain; play in the
>>> corner, edges, and center would all be more complicated patterns, and
>>> there would be far more good candidates to consider at each ply and
>>> rollouts would be much less stable and less accurate.
>>>
>>

> yes.
>>
>> the normal board size is 19x19 because the amount of territory in the
>> sides and corners is slightly larger than the amount of territory in the
>> middle.
>>
>> thanks
>>
>> --
>> Honesty is a very expensive gift. So, don't expect it from cheap people -
>> Warren Buffett
>> http://tayek.com/
>>
>>
>> ___
>> Computer-go mailing list
>>
> ___
> Computer-go mailing list
___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go

Re: [Computer-go] Would a larger board (25x25) dramatically reduce AlphaGos skill?

[Yuancheng Luo]

Conv net should be robust. From image processing domian, these are feature
detectors (shape in case of go) that are invariant to translations (moving
shape left right up down along board). Enlarging board wouldnt put bot at
 disadvantage in evaluating local positions.


On Tuesday, March 22, 2016, Ray Tayek  wrote:

> On 3/22/2016 5:21 PM, Lukas van de Wiel wrote:
>
> It would reduce Alphago, because there is less training material in the
> form of high-dan-games, to train the policy network.
>
> It would also reduce the skill of a human opponent, because (s)he would
> have less experience on a larger board, just as AlphaGo.
>
> It would be fun to see which can adapt better.
>
>
> human would adapt quickly after a few games (say 10 or so).
>
> thanks
>
>
> On Wed, Mar 23, 2016 at 1:18 PM, Ray Tayek  > wrote:
>
>> On 3/22/2016 11:25 AM, Tom M wrote:
>>
>>> I suspect that even with a similarly large training sample for
>>> initialization that AlphaGo would suffer a major reduction in apparent
>>> skill level.
>>>
>> i think a human would also.
>>
>>>The CNN would require many more layers of convolution;
>>> the valuation of positions would be much more uncertain; play in the
>>> corner, edges, and center would all be more complicated patterns, and
>>> there would be far more good candidates to consider at each ply and
>>> rollouts would be much less stable and less accurate.
>>>
>> yes.
>>
>> the normal board size is 19x19 because the amount of territory in the
>> sides and corners is slightly larger than the amount of territory in the
>> middle.
>>
>> thanks
>>
>> --
>> Honesty is a very expensive gift. So, don't expect it from cheap people -
>> Warren Buffett
>> http://tayek.com/
>>
>>
>> ___
>> Computer-go mailing list
>> Computer-go@computer-go.org
>> 
>> http://computer-go.org/mailman/listinfo/computer-go
>>
>
>
>
> ___
> Computer-go mailing listcomputer...@computer-go.org 
> http://computer-go.org/mailman/listinfo/computer-go
>
>
>
> --
> Honesty is a very expensive gift. So, don't expect it from cheap people - 
> Warren Buffetthttp://tayek.com/
>
>

-- 
Yuancheng [Mike] Luo
www.umiacs.umd.edu/~yluo1 
___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go

Re: [Computer-go] Would a larger board (25x25) dramatically reduce AlphaGos skill?

[Ray Tayek]

On 3/22/2016 5:21 PM, Lukas van de Wiel wrote:
It would reduce Alphago, because there is less training material in 
the form of high-dan-games, to train the policy network.


It would also reduce the skill of a human opponent, because (s)he 
would have less experience on a larger board, just as AlphaGo.


It would be fun to see which can adapt better.


human would adapt quickly after a few games (say 10 or so).

thanks



On Wed, Mar 23, 2016 at 1:18 PM, Ray Tayek > wrote:


On 3/22/2016 11:25 AM, Tom M wrote:

I suspect that even with a similarly large training sample for
initialization that AlphaGo would suffer a major reduction in
apparent
skill level.

i think a human would also.

   The CNN would require many more layers of convolution;
the valuation of positions would be much more uncertain; play
in the
corner, edges, and center would all be more complicated
patterns, and
there would be far more good candidates to consider at each
ply and
rollouts would be much less stable and less accurate.

yes.

the normal board size is 19x19 because the amount of territory in
the sides and corners is slightly larger than the amount of
territory in the middle.

thanks

-- 
Honesty is a very expensive gift. So, don't expect it from cheap

people - Warren Buffett
http://tayek.com/


___
Computer-go mailing list
Computer-go@computer-go.org 
http://computer-go.org/mailman/listinfo/computer-go




___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go



--
Honesty is a very expensive gift. So, don't expect it from cheap people - 
Warren Buffett
http://tayek.com/

___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go

Re: [Computer-go] Would a larger board (25x25) dramatically reduce AlphaGos skill?

[Lukas van de Wiel]

It would reduce Alphago, because there is less training material in the
form of high-dan-games, to train the policy network.

It would also reduce the skill of a human opponent, because (s)he would
have less experience on a larger board, just as AlphaGo.

It would be fun to see which can adapt better.

Cheers
Lukas

On Wed, Mar 23, 2016 at 1:18 PM, Ray Tayek  wrote:

> On 3/22/2016 11:25 AM, Tom M wrote:
>
>> I suspect that even with a similarly large training sample for
>> initialization that AlphaGo would suffer a major reduction in apparent
>> skill level.
>>
> i think a human would also.
>
>>The CNN would require many more layers of convolution;
>> the valuation of positions would be much more uncertain; play in the
>> corner, edges, and center would all be more complicated patterns, and
>> there would be far more good candidates to consider at each ply and
>> rollouts would be much less stable and less accurate.
>>
> yes.
>
> the normal board size is 19x19 because the amount of territory in the
> sides and corners is slightly larger than the amount of territory in the
> middle.
>
> thanks
>
> --
> Honesty is a very expensive gift. So, don't expect it from cheap people -
> Warren Buffett
> http://tayek.com/
>
>
> ___
> Computer-go mailing list
> Computer-go@computer-go.org
> http://computer-go.org/mailman/listinfo/computer-go
>
___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go

Re: [Computer-go] Would a larger board (25x25) dramatically reduce AlphaGos skill?

[Ray Tayek]

On 3/22/2016 11:25 AM, Tom M wrote:

I suspect that even with a similarly large training sample for
initialization that AlphaGo would suffer a major reduction in apparent
skill level.

i think a human would also.

   The CNN would require many more layers of convolution;
the valuation of positions would be much more uncertain; play in the
corner, edges, and center would all be more complicated patterns, and
there would be far more good candidates to consider at each ply and
rollouts would be much less stable and less accurate.

yes.

the normal board size is 19x19 because the amount of territory in the 
sides and corners is slightly larger than the amount of territory in the 
middle.


thanks

--
Honesty is a very expensive gift. So, don't expect it from cheap people - 
Warren Buffett
http://tayek.com/

___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go

Re: [Computer-go] Would a larger board (25x25) dramatically reduce AlphaGos skill?

[uurtamo .]

Ko is what makes this game difficult, from a theoretical point of view.

I suspect ko+unresolved groups is where it's at.

s.
On Mar 22, 2016 11:25 AM, "Tom M"  wrote:

> I suspect that even with a similarly large training sample for
> initialization that AlphaGo would suffer a major reduction in apparent
> skill level.  The CNN would require many more layers of convolution;
> the valuation of positions would be much more uncertain; play in the
> corner, edges, and center would all be more complicated patterns, and
> there would be far more good candidates to consider at each ply and
> rollouts would be much less stable and less accurate.
> ___
> Computer-go mailing list
> Computer-go@computer-go.org
> http://computer-go.org/mailman/listinfo/computer-go
___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go

[Computer-go] Would a larger board (25x25) dramatically reduce AlphaGos skill?

[Tom M]

I suspect that even with a similarly large training sample for
initialization that AlphaGo would suffer a major reduction in apparent
skill level.  The CNN would require many more layers of convolution;
the valuation of positions would be much more uncertain; play in the
corner, edges, and center would all be more complicated patterns, and
there would be far more good candidates to consider at each ply and
rollouts would be much less stable and less accurate.
___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go