Anyone gotten facebook's OpenGo to compile yet?

[hydrogenpi7]

How much stronger is it than leela zero on only one playout?


https://github.com/gcp/leela-zero/issues/1311

https://github.com/pytorch/ELF/issues/1

[macelee]

How much stronger is it than leela zero on only one playout?

According to a news article, 200-0 against LZ and 15-0 against top Korean pros.

[Uberdude]

Edit: I moved my general comments here so this thread can focus on compilation help.

[Sneegurd]

You can play it with this Leela fork:
https://github.com/Ka-zam/leela-zero/releases

Use this converted Elf net:
http://zero.sjeng.org/networks/62b5417b ... 97df4b9.gz

[Gomoto]

niiiiice!!!!!!!!!!!!!!!!!

Having fun with this combo.

After the download of the two components I copied them into a Lizzie directory and it is working.

I am already updating some of my standard josekis ...

(It is not as stable as regular leelaz, but it is working. It does not like if I enter my moves and variations too fast. I have to enter them gentle, difficult if you are exited )

[Gomoto]

And the invincible shusaku was right after all:

joseki.jpg (80.73 KiB) Viewed 12871 times

But only with the shusaku kosumi, in the fuseki he should have played probably:

shusaku.jpg (79.46 KiB) Viewed 12870 times

[Gomoto]

Now Haylee can win against Leela Zero, she only has to train the remaining few days very hard with LeelaElfLizzie .

[Gomoto]

It is much more stable with Lizzie 0.3 on my machine.

[Gomoto]

Elf playing versus Leela Zero on my box:

viewtopic.php?p=230789#p230789

[Sneegurd]

Gomoto, Lizzie shouldn't be able to use the LZ compile yet (the new one which is able to use the FB weights). I assume what you show is the Lizzie output of classical LZ weights.
Please look here: https://www.reddit.com/r/cbaduk/comment ... rk_for_me/

[Gomoto]

I am using Lizzie 0.3 with

You can play it with this Leela fork:
https://github.com/Ka-zam/leela-zero/releases

Use this converted Elf net:
http://zero.sjeng.org/networks/62b5417b ... 97df4b9.gz

and this works fine on my box.

If your link to the converted Elf net is real, I show real Elf Lizzie 0.3 output indeed

And by the way the, converted Elf net Leela Combo is beating Leela Zero quite easily here on my machine, so I think it is quite real.



(Lizzie 0.4 does stop pondering all the time, but it works for a few moves as well.)

[dfan]

Gomoto, Lizzie shouldn't be able to use the LZ compile yet (the new one which is able to use the FB weights).

I am running Lizzie with the latest Leela Zero with ELF weights with no problem. You do still have to patch Leela Zero to output ponder information on the fly.

[hydrogenpi7]

Okay so I'm trying to think this out logically. Facebook releases ELF OpenGo, instantly becoming by far the strongest public open source /weights Go AI program, overnight gets adopted as the baseline by many other programs.... so going forward most programs are going to be more or less the same strength if not simply identical whitelabel altogether, so what is left? Seems to be the "engine" part is now more or less solved and even I would say undifferentiatable. superhuman on a gtx 970, is basically the end of the road.

The remaining things of innovation are GUI, analysis, high handicap, different komi, teaching tools, etc etc

And marketing/branding/mindshare/PR for the Go bots (including Leela Zero) will perhaps be more important than ever before. We gonna see a consolidation of Go AI bots and my guess is only one or two will survive this, and that is if they are lucky.

Some immediate implications is that essentially its killed commerical Go at least from the standpoint of selling engines go. We can't compare to Chess because not only does Chess have an order of magnitude larger userbase esp in the West, Chess also enjoyed a good two decades whereby classical algorithms and programming made it such that there was a healthy ecosystem of different engines completing with one another for top listings. WIth the advent of the "zero" method, all zero programs converge to the same ultimate state and its just a matter of compute. There is really nothing left to do. More or less.

This also means there is little to no more point in having Go AI engine competition and matches. We already see cgos is defunt and its benchmark is less and less useful, UEC cup ended, Zen pulled the plug and called it quits, I seriously doubt we'll see another version or edition of CrazyStone, and now with so many engines adopting the facebook weights, whats the point? I see this as portending the demise of Go AI competitions and engine vs engine games as well. Think about it, LZ had beat DolBaram in that last competition match, now DolBaram adopts ELF weights, and ELF is stronger than both Pheonix and FineArt... it doesn't take much to put two and two together and see where this is headed... Didn't Golaxy just beat Ke Jie last week? Ill bet that was the shortest triump ever. And whatever aire of exclusitivity that FineArt enjoyed prior to the facebook event has now been obliterated, top pro in China no longer need to use FineArt to get competitive advantage in training when everyone in the world on half a decent graphics card can now run the same or better. The implications are indeed far fetching.

Lets examine the distributed community based crowd computing aspect angle. It took the public six months to get LZ to top pro level from scratch and yet facebook only needed two weeks and argueably far surpassed top pro levels and went deep into superhuman arena. Not that I know it is going to happen, but there is nothing to prevent facebook from doing it again, say another couple months down the road it can sudden drop a new weight that will be the new state of the art and far surpassing anything any community effort could have hoped to come up with within that allocation of time. Who knows maybe Google will see all this and publish the AGZ weights, or maybe in another few months the second round of weights that facebook puts out will far suprass AGZ altogether! In light of the recent developments these are all realistic possibilities now! But none of these possibilities foster morale for community initatives.

I'm thankful that prior to facebook dropping ELF onto the world, that LZ already reached and imho surpassed top pro level on its last/final network 131, (I see 132 just came out hours after the Haylee game 2 and is 60% stronger!) and that LZ project was able to convert ELF weights into native LZ format so that it can be used just like any other weightfile and now its even working great in Lizzie.

I hope that Leela Zero project finds a way to position itself to best take advantage of this new and changing landscape. By far it enjoys the most mindshare in the community of Go at large right now and I hope it continues to evolve and find ways of remaining relevant and bringing value to people's lives.

[Gomoto]

There is still enough headroom for engine development. Elf weights are also still changing sometimes all of a sudden after a move is made.

I think it is quite interesting to watch different training runs and have a look if the same variations and play style emerges or if there is room for different play styles. I dont think it would be a good idea if everybody is training on the same set of weights, let us keep them seperated for a while.

I enjoy watching Elf versus Leela Zero weights, also it is a little one sided at the current point in time.

If neural networks are indeed a model for human intuition, this is a indicator there will develop several different but competitive playstyles. This is a good thing for go. Let us keep calm, go will not develop into tic-tac-toe, because one can not brute force it.

[pnprog]

[...]
I hope that Leela Zero project finds a way to position itself to best take advantage of this new and changing landscape. By far it enjoys the most mindshare in the community of Go at large right now and I hope it continues to evolve and find ways of remaining relevant and bringing value to people's lives.

You may it sounds like they are companies struggling for market share

Sometime, the journey is more important that the destination. Since AlphaGo was unveiled in 2016, we have been moving at fast speed into a new Go era. It may seems a lot of energy has been, and is being wasted along the road, but I am confident most of us are enjoying the journey anyway. I hope you can enjoy as well!