LCZero update

[jhellis3]

[CMCanavessi]

So from Gen 6 to Gen 8, LCZero got +190 elo, in 2 days. Imagine if the project catches up and more people help train it. It would blow our minds. I believe that Gen 10-12 will already be around 1000 elo, by the weekend.

Like most of the ppl you don't quite understand how DCNN training works.
Sooner than later you'll hit a plateau and the training will saturate. The higher you go, the harder it becomes to improve DCNN.

I know that gaining strenght it will slow down considerably, eventually. But it's still too soon for that. We'll see, but until it hits ~2500 elo i believe it will grow quite fast.

[Jhoravi]

Here's a "ready to use" package for anyone that wants to help train LCZero. Just unpack the .7z file in a folder and run the file Setup.bat
It will ask for a username and password for the 1st time, and then it will automatically create a Train.bat file and start training.

When you want to run it, just double click the file Train.bat and all should work well.

http://www.mediafire.com/file/0dxx0k6f2y5bw5l/LCZero.7z

It's already on Generation 8, and it's looking better every day. Gaining elo fast.

Thank you brother! It works like a charm! But it made me realize that my laptop is just averaging 170 nps . But I'm sure a lot of my facebook friends have fast PC. I wish somebody can make a neat graphical interface so that those common people can participate.

[jkiliani]

So from Gen 6 to Gen 8, LCZero got +190 elo, in 2 days. Imagine if the project catches up and more people help train it. It would blow our minds. I believe that Gen 10-12 will already be around 1000 elo, by the weekend.

Like most of the ppl you don't quite understand how DCNN training works.
Sooner than later you'll hit a plateau and the training will saturate. The higher you go, the harder it becomes to improve DCNN.

While you are correct in principle, you're overlooking that there are ways to deal with that, as Leela Zero already demonstrated. Once you reach a plateau, you can simply use the self-play games you already have to bootstrap a larger neural net, which will usually achieve a significant initial jump and be able to train to a higher level. Once you stall again, rinse and repeat.

[jkiliani]

I know that gaining strenght it will slow down considerably, eventually. But it's still too soon for that. We'll see, but until it hits ~2500 elo i believe it will grow quite fast.

My best guess is that the current 6 block, 64 filters net will be able to get to around 2000-2200 elo. At that point I believe the first bootstrap is due, maybe to 10 block, 128 filters. But we will see where the learning curve starts to flatten...

[Milos]

So from Gen 6 to Gen 8, LCZero got +190 elo, in 2 days. Imagine if the project catches up and more people help train it. It would blow our minds. I believe that Gen 10-12 will already be around 1000 elo, by the weekend.

Like most of the ppl you don't quite understand how DCNN training works.
Sooner than later you'll hit a plateau and the training will saturate. The higher you go, the harder it becomes to improve DCNN.

While you are correct in principle, you're overlooking that there are ways to deal with that, as Leela Zero already demonstrated. Once you reach a plateau, you can simply use the self-play games you already have to bootstrap a larger neural net, which will usually achieve a significant initial jump and be able to train to a higher level. Once you stall again, rinse and repeat.

In some cases larger net would help, in others not, which is most probably the case with A0.
Training larger net requires much more resources for self-play games. In case of LCZero, they are already struggling with relatively small net as it is now to get a decent number of games.
So enthusiasm regarding LCZero is pretty much in vain.

[jkiliani]

So from Gen 6 to Gen 8, LCZero got +190 elo, in 2 days. Imagine if the project catches up and more people help train it. It would blow our minds. I believe that Gen 10-12 will already be around 1000 elo, by the weekend.

Like most of the ppl you don't quite understand how DCNN training works.
Sooner than later you'll hit a plateau and the training will saturate. The higher you go, the harder it becomes to improve DCNN.

While you are correct in principle, you're overlooking that there are ways to deal with that, as Leela Zero already demonstrated. Once you reach a plateau, you can simply use the self-play games you already have to bootstrap a larger neural net, which will usually achieve a significant initial jump and be able to train to a higher level. Once you stall again, rinse and repeat.

In some cases larger net would help, in others not, which is most probably the case with A0.
Training larger net requires much more resources for self-play games. In case of LCZero, they are already struggling with relatively small net as it is now to get a decent number of games.
So enthusiasm regarding LCZero is pretty much in vain.

A larger net always helps, since more weights can get better representation of the search output, and a ResNet also doesn't suffer from vanishing gradients anymore like earlier network architectures. Leela Zero got a big boost with every network expansion, and you can't tell me that somehow chess is so fundamentally different that the same wouldn't apply here.

Sure, self-play speed will go down with every network expansion, but it will also go up a lot as the project gains traction and more people contribute. The stronger it gets, the more publicity will there be.

Enthusiasm for LCZero is very well founded, and shared by everyone who read the Deepmind papers and understood them at least at some level. No-one is forcing you to contribute, feel free to watch the downfall of the Alpha-Beta engines from the sidelines.

[Leo]

Its logical to try LCZero to see if it works. I am skeptical but not a pessimist.

[Uri Blass]

Its logical to try LCZero to see if it works. I am skeptical but not a pessimist.

I look at the games and I see a lot of stupid one ply blunder that lose material.
I do not know what they do but if after many thousands of games it plays like that then I do not believe in it.

[CMCanavessi]

Its logical to try LCZero to see if it works. I am skeptical but not a pessimist.

I look at the games and I see a lot of stupid one ply blunder that lose material.
I do not know what they do but if after many thousands of games it plays like that then I do not believe in it.

Of course it will make mistakes like that, the learning phase started 1 week ago. You can't expect it to become the new stockfish in 1 week. It has gained like 700 elo since it started training...