Ok, so the idea is to replace the search with a single playout? How is that a 'pure network'?MonteCarlo wrote:Yes, he's talking about a 1 visit "search", as he explicitly said (and I highlighted above).Albert Silver wrote:It is possible because you presented a match result, then quoted LCZero's strength at ~2900 CCRL, which I am pretty sure is not 2000 Elo. I also have no idea what the strength of a 'pure network' even means. If you are suggesting plain single node play with no search tree of any kind then I am sure it is MUCH weaker than 2000 Elo.Dariusz Orzechowski wrote:I have no idea how it's possible to spin it this way. He quoted what I said about pure network and answered with a statement that is demonstrably not true.Dann Corbit wrote:I think this is apples and oranges.Dariusz Orzechowski wrote:This is just highly innacurate. Pure network is perhaps around 2000 now what makes it a very decent player anyway.Albert Silver wrote:I think it is already at least SuperGM level.Dariusz Orzechowski wrote:For me it's exciting but it's not that incredible as I was following Leela Zero in go and something similar happened there. I like playing against pure policy networks (1 playout, no search), it's a lot of fun in both chess and go. In go pure network is already around 3-4 dan level which would correspond to FM/IM level in chess I guess. I'm pretty confident it's possible to get pure network to around 5-6 dan in go and GM level in chess.
Albert was talking about the current state of the LCZero project and you about using a pure network.
In classical time controls, sure, at 1 visit LC0 is probably not 2000 on any human scale. However, at any time control where a human has to even remotely try to match the pace of a 1 visit "search", it's likely far stronger than 2000
LCZero Accomplishments and Goals Thus Far
Moderators: hgm, Rebel, chrisw
-
- Posts: 3019
- Joined: Wed Mar 08, 2006 9:57 pm
- Location: Rio de Janeiro, Brazil
Re: LCZero Accomplishments and Goals Thus Far
"Tactics are the bricks and sticks that make up a game, but positional play is the architectural blueprint."
-
- Posts: 188
- Joined: Sun Dec 25, 2016 4:59 pm
Re: LCZero Accomplishments and Goals Thus Far
Well, the search algorithm plays the move with the most visits.
The initial visit is made to the move with the highest probability from the policy head.
With a 1 visit "search", the search will evaluate the position after the move with the highest probability from the policy head, but no matter what the evaluation is, it will play that move.
It has 1 visit, and everything else has 0.
So you're literally just playing against the policy head, as he indicated he liked to do
The initial visit is made to the move with the highest probability from the policy head.
With a 1 visit "search", the search will evaluate the position after the move with the highest probability from the policy head, but no matter what the evaluation is, it will play that move.
It has 1 visit, and everything else has 0.
So you're literally just playing against the policy head, as he indicated he liked to do
-
- Posts: 3019
- Joined: Wed Mar 08, 2006 9:57 pm
- Location: Rio de Janeiro, Brazil
Re: LCZero Accomplishments and Goals Thus Far
So it is essentially just evaluating the position with the highest value move according to its policy. Ok, well, needless to say, if you reduce the time to the human to impossible controls such as one minute or less with no increment, the human will lose in all likelihood at some juncture, but if that is qualifying the 'pure network' as 2000 or GM, we can say it is already a GM at g/10s. On the other hand, if you set the lower limit to g/5 (minutes) and use this 'pure network', it will never reach GM level, due to the tactics. Unless you handpick the GM I suppose...MonteCarlo wrote:Well, the search algorithm plays the move with the most visits.
The initial visit is made to the move with the highest probability from the policy head.
With a 1 visit "search", the search will evaluate the position after the move with the highest probability from the policy head, but no matter what the evaluation is, it will play that move.
It has 1 visit, and everything else has 0.
So you're literally just playing against the policy head, as he indicated he liked to do
"Tactics are the bricks and sticks that make up a game, but positional play is the architectural blueprint."
-
- Posts: 188
- Joined: Sun Dec 25, 2016 4:59 pm
Re: LCZero Accomplishments and Goals Thus Far
On the time control, even if it's 1+1, I'd wager 1 visit is around 2000 now.Albert Silver wrote:So it is essentially just evaluating the position with the highest value move according to its policy. Ok, well, needless to say, if you reduce the time to the human to impossible controls such as one minute or less with no increment, the human will lose in all likelihood at some juncture, but if that is qualifying the 'pure network' as 2000 or GM, we can say it is already a GM at g/10s. On the other hand, if you set the lower limit to g/5 (minutes) and use this 'pure network', it will never reach GM level, due to the tactics. Unless you handpick the GM I suppose...MonteCarlo wrote:Well, the search algorithm plays the move with the most visits.
The initial visit is made to the move with the highest probability from the policy head.
With a 1 visit "search", the search will evaluate the position after the move with the highest probability from the policy head, but no matter what the evaluation is, it will play that move.
It has 1 visit, and everything else has 0.
So you're literally just playing against the policy head, as he indicated he liked to do
On the claim that 1 visit will never be GM-level at G/5 or slower, that should age well Let's talk again in a year
-
- Posts: 3019
- Joined: Wed Mar 08, 2006 9:57 pm
- Location: Rio de Janeiro, Brazil
Re: LCZero Accomplishments and Goals Thus Far
At 1+1 I would imagine it could do a lot better than 2000.MonteCarlo wrote:On the time control, even if it's 1+1, I'd wager 1 visit is around 2000 now.Albert Silver wrote:So it is essentially just evaluating the position with the highest value move according to its policy. Ok, well, needless to say, if you reduce the time to the human to impossible controls such as one minute or less with no increment, the human will lose in all likelihood at some juncture, but if that is qualifying the 'pure network' as 2000 or GM, we can say it is already a GM at g/10s. On the other hand, if you set the lower limit to g/5 (minutes) and use this 'pure network', it will never reach GM level, due to the tactics. Unless you handpick the GM I suppose...MonteCarlo wrote:Well, the search algorithm plays the move with the most visits.
The initial visit is made to the move with the highest probability from the policy head.
With a 1 visit "search", the search will evaluate the position after the move with the highest probability from the policy head, but no matter what the evaluation is, it will play that move.
It has 1 visit, and everything else has 0.
So you're literally just playing against the policy head, as he indicated he liked to do
On the claim that 1 visit will never be GM-level at G/5 or slower, that should age well Let's talk again in a year
"Tactics are the bricks and sticks that make up a game, but positional play is the architectural blueprint."
-
- Posts: 13447
- Joined: Wed Mar 08, 2006 9:02 pm
- Location: Dallas, Texas
- Full name: Matthew Hull
Re: LCZero Accomplishments and Goals Thus Far
Your demand for uniform platform comparison is commutative. Why not demand all the other engines run on a GPU?Daniel Shawul wrote:Sigh..wake me up when it is 2800 elo running on singe CPU core, which is what every other engine uses in rating lists. As far as I am concerned, it is still a 2100 elo engine there.
Then it would be "equal".
Matthew Hull
-
- Posts: 1470
- Joined: Mon Apr 23, 2018 7:54 am
Re: LCZero Accomplishments and Goals Thus Far
So if I teach my pet frog to play chess, I should demand you play it underwater rather than on land, because demands should be "commutative"?mhull wrote:Your demand for uniform platform comparison is commutative. Why not demand all the other engines run on a GPU?Daniel Shawul wrote:Sigh..wake me up when it is 2800 elo running on singe CPU core, which is what every other engine uses in rating lists. As far as I am concerned, it is still a 2100 elo engine there.
Then it would be "equal".
-
- Posts: 1346
- Joined: Sat Apr 19, 2014 1:47 pm
Re: LCZero Accomplishments and Goals Thus Far
I guess someone is frustrated to have a computer with 2 core and no graphic card and instead of saying "I d like so much to play with a strong LCzero" he just says "LCzero is bad with 1 core".
Okey, try meditation. It might help.
Okey, try meditation. It might help.
-
- Posts: 27811
- Joined: Fri Mar 10, 2006 10:06 am
- Location: Amsterdam
- Full name: H G Muller
Re: LCZero Accomplishments and Goals Thus Far
A better analogy: If I teach my pet shark to play chess, wouldn't it be a bit unfair to insist he will play it sitting in a chair in a room filled with air?jp wrote:So if I teach my pet frog to play chess, I should demand you play it underwater rather than on land, because demands should be "commutative"?
-
- Posts: 1470
- Joined: Mon Apr 23, 2018 7:54 am
Re: LCZero Accomplishments and Goals Thus Far
My frog can survive on land even if he prefers the pond. Your shark would just die.hgm wrote:A better analogy: If I teach my pet shark to play chess, wouldn't it be a bit unfair to insist he will play it sitting in a chair in a room filled with air?jp wrote:So if I teach my pet frog to play chess, I should demand you play it underwater rather than on land, because demands should be "commutative"?
You jumping into the pond with my frog is also fairer than me being forced into your shark's jaws in the ocean.