Re: [Computer-go] Google used TPUs for AlphaGo

Body of the Japanese text says "TPU is an ASIC developed for Deep Learning with 'ten' times better performance per Watt compared to other technologies such as GPU or FPGA, according to Pichai." So may be it is easier to consider "10 times over Stratix(1) or Vertex7/Xilinx(2)" and not 13 times over

[Computer-go] some info for you

Hello, Isn't it the information you were looking for? Read about it here Typos courtesy of my iPhone, Christy Simmons ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.or

Re: [Computer-go] Question on TPU hardware

I don’t expect AlphaGo will be available at any price, but I expect similar strength programs will be running on high end PCs in a few years. The AlphaGo team has done an outstanding job of exploring the solution space and showing us the way. Others can now tweak and optimize and find more effi

Re: [Computer-go] Commonly used neural network architectures

Thanks David! You’re right, it should be possible to somehow read out that prototxt file. I’ll see how that goes for me. :) The rest of your info is still a bit mysterious for me. But I will figure it out eventually. Urban On Fri, May 20, 2016 at 9:49 PM, David Fotland wrote: > The alphaGo netw

Re: [Computer-go] Question on TPU hardware

On Sat, May 21, 2016 at 09:25:53AM +0200, "Ingo Althöfer" wrote: > I have a question (maybe to Aja): > > Concerning the TPU hardware used in AlphaGo, how long will it take > until that system (all together, including Go software) will be > available for 20,000 Euro or less? I don't have an answe

[Computer-go] Question on TPU hardware

I have a question (maybe to Aja): Concerning the TPU hardware used in AlphaGo, how long will it take until that system (all together, including Go software) will be available for 20,000 Euro or less? Ingo. ___ Computer-go mailing list Computer-go@compu

Re: [Computer-go] Google used TPUs for AlphaGo

> http://itpro.nikkeibp.co.jp/atcl/column/15/061500148/051900060/ > (in Japanese). The performance/watt is about 13 times better, > a photo in the article shows. Has anyone found out exactly what the "Other" in the photo is? The Google blog was also rather vague on this. (If you didn't click t