[Computer-go] Zero performance
soring at gmail.com
Fri Oct 20 13:41:15 PDT 2017
Training of AlphaGo Zero has been done on thousands of TPUs, according to
Maybe that should explain the difference in orders of magnitude that you
On Fri, Oct 20, 2017 at 10:44 AM, Gian-Carlo Pascutto <gcp at sjeng.org> wrote:
> I reconstructed the full AlphaGo Zero network in Caffe:
> I did some performance measurements, with what should be
> state-of-the-art on consumer hardware:
> GTX 1080 Ti
> NVIDIA-Caffe + CUDA 9 + cuDNN 7
> batch size = 8
> Memory use is about ~2G. (It's much more for learning, the original
> minibatch size of 32 wouldn't fit on this card!)
> Running 2000 iterations takes 93 seconds.
> In the AlphaGo paper, they claim 0.4 seconds to do 1600 MCTS
> simulations, and they expand 1 node per visit (if I got it right) so
> that would be 1600 network evaluations as well, or 200 of my iterations.
> So it would take me ~9.3s to produce a self-play move, compared to 0.4s
> for them.
> I would like to extrapolate how long it will take to reproduce the
> research, but I think I'm missing how many GPUs are in each self-play
> worker (4 TPU or 64 GPU or ?), or perhaps the average length of the games.
> Let's say the latter is around 200 moves. They generated 29 million
> games for the final result, which means it's going to take me about 1700
> years to replicate this. I initially estimated 7 years based on the
> reported 64 GPU vs 1 GPU, but this seems far worse. Did I miss anything
> in the calculations above, or was it really a *pile* of those 64 GPU
> Because the performance on playing seems reasonable (you would be able
> to actually run the MCTS on a consumer machine, and hence end up with a
> strong program), I would be interested in setting up a distributed
> effort for this. But realistically there will be maybe 10 people
> joining, 80 if we're very lucky (looking at Stockfish numbers). That
> means it'd still take 20 to 170 years.
> Someone please tell me I missed a factor of 100 or more somewhere. I'd
> love to be wrong here.
> Computer-go mailing list
> Computer-go at computer-go.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Computer-go