5 ms·

There is a public distributed effort happening for Go right now: http://zero.sjeng.org/ http://zero.sjeng.org/. They've been doing a fantastic job, and just re

There is a public distributed effort happening for Go right now: http://zero.sjeng.org/ http://zero.sjeng.org/. They've been doing a fantastic job, and just recently fixed a big training bug that has resulted in a large strength increase. I ported over from GCP's Go implementation to chess: https://github.com/glinscott/leela-chess https://github.com/glinscott/leela-chess. The distributed part isn't ready to go yet, we are still working the bugs out using supervised training, but will be launching soon!

yazr 9y ago
From the leela-zero page : > If your CPU is not very recent (Haswell or newer, Ryzen or newer), performance will be outright bad, Obviously TPU >> GPU >> CPU But are there special vector instructions added in Haswell? Or is this just general preference for a multi core newish cpu ?
- cbuq 9y ago
  Not sure if this is their reasoning, but Haswell introduced AVX2 instructions (https://en.wikipedia.org/wiki/Advanced_Vector_Extensions#Advanced_Vector_Extensions_2 https://en.wikipedia.org/wiki/Advanced_Vector_Extensions#Adv...)
- wohlergehen 9y ago
  AVX2 introduces vectorized Fused-Multiply-Add, for a 2x boost in linear algebra ops.
grondilu 9y ago
Regarding supervised learning, what data do you use? Would it make sense to use endgame tablebases[1]? 1. https://en.wikipedia.org/wiki/Endgame_tablebase https://en.wikipedia.org/wiki/Endgame_tablebase
- glinscott 9y ago
  We are using data from both human grandmaster games and self-play games of a recent Stockfish version. Both have resulted in networks that play reasonable openings, but we had some issues with the value head not understanding good positions. We think we have a line on why this is happening (too few weights in the final stage of the network), but this is exactly the purpose of the supervised learning debugging phase :).
NelsonMinar 9y ago
I've been having fun following along with Leela Zero, it's a great way to understand how a project like this goes at significant scale. Good luck with Leela Chess, I'm excited for it!
sprt 9y ago
What does SPRT stand for in that context?
- tux3 9y ago
  Sequential probability ratio test, it is essentially a test that distinguishes between two hypotheses with high probability. In the case of Leela Zero the idea is to train new networks continuously and have them fight against the current best network, which is replaced only when a new network is statistically stronger, according to SPRT.
iandanforth 9y ago
This is really cool! The chart on that page makes it look like Leela Zero is already much much better than AlphaZero (~7400 Elo vs ~5200 Elo). I suspect I'm misinterpreting something though, could you clarify?
- tux3 9y ago
  Leela Zero's ELO graph assumes that 0 ELO is completely random play, as a simple reference point. On the other hand Alphago uses the more common ELO scale where 0 is roughly equivalent to a beginner who knows the rules, so you can't directly compare the two.
adw 9y ago
The irony here being you're using Stockfish move representation in a codebase written by the author of Sjeng :-)

yazr 9y ago

From the leela-zero page : > If your CPU is not very recent (Haswell or newer, Ryzen or newer), performance will be outright bad, Obviously TPU >> GPU >> CPU But are there special vector instructions added in Haswell? Or is this just general preference for a multi core newish cpu ?

cbuq 9y ago
Not sure if this is their reasoning, but Haswell introduced AVX2 instructions (https://en.wikipedia.org/wiki/Advanced_Vector_Extensions#Advanced_Vector_Extensions_2 https://en.wikipedia.org/wiki/Advanced_Vector_Extensions#Adv...)
wohlergehen 9y ago
AVX2 introduces vectorized Fused-Multiply-Add, for a 2x boost in linear algebra ops.

grondilu 9y ago

Regarding supervised learning, what data do you use? Would it make sense to use endgame tablebases[1]? 1. https://en.wikipedia.org/wiki/Endgame_tablebase https://en.wikipedia.org/wiki/Endgame_tablebase

glinscott 9y ago
We are using data from both human grandmaster games and self-play games of a recent Stockfish version. Both have resulted in networks that play reasonable openings, but we had some issues with the value head not understanding good positions. We think we have a line on why this is happening (too few weights in the final stage of the network), but this is exactly the purpose of the supervised learning debugging phase :).

NelsonMinar 9y ago

I've been having fun following along with Leela Zero, it's a great way to understand how a project like this goes at significant scale. Good luck with Leela Chess, I'm excited for it!

sprt 9y ago

What does SPRT stand for in that context?

tux3 9y ago
Sequential probability ratio test, it is essentially a test that distinguishes between two hypotheses with high probability. In the case of Leela Zero the idea is to train new networks continuously and have them fight against the current best network, which is replaced only when a new network is statistically stronger, according to SPRT.

iandanforth 9y ago

This is really cool! The chart on that page makes it look like Leela Zero is already much much better than AlphaZero (~7400 Elo vs ~5200 Elo). I suspect I'm misinterpreting something though, could you clarify?

tux3 9y ago
Leela Zero's ELO graph assumes that 0 ELO is completely random play, as a simple reference point. On the other hand Alphago uses the more common ELO scale where 0 is roughly equivalent to a beginner who knows the rules, so you can't directly compare the two.

adw 9y ago

The irony here being you're using Stockfish move representation in a codebase written by the author of Sjeng :-)