[Computer-go] Using GPUs?
nikpapa at gmail.com
Sat Jun 27 04:55:36 PDT 2015
On Fri, Jun 26, 2015 at 4:20 PM, Darren Cook <darren at dcook.org> wrote:
> Confirmed here:
> So, currently they use a 32-bit float, rather than a 64-bit double, but
> will reduce that to 16-bit to get a double speed-up. Assuming they've
> been listening to customers properly, that must mean 16-bit floats are
> good enough for neural nets?
> Apparently you can use 16-bit representations in DNNs with little or no
degradation in accuracy:
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Computer-go