Re: GPU, neural networks and APU
Karen Shaeffer <shaeffer-IwRZ8fqS3AoztatW0fm/[email protected]>
| Newsgroups | gmane.org.user-groups.linux.svlug |
|---|---|
| Message-ID | <[email protected]> |
On Fri, Aug 05, 2016 at 06:49:40AM -0700, Mehma Sarja wrote: > This is off subject, but if you in are into math, Peter Woit's "Not Even > Wrong" is a good read on quantum chromodynamics and string theory. > > Yudhvir Hi Yudhvir, Mathematics is not off topic in my opinion. In practice mathematics are applied on computational computer networks that almost always are running on a linux distribution and implemented using open source libraries. I've returned to my roots and am now fully focused on mathematics and algorithm development these days. But I remain intensely interested in linux and open source libraries, because to actually use that math, I run a linux network and use a significant number of open source libraries. And so mathematics are abstracted from a direct discussion of linux, but they are very relevant -- driving the vast majority of new applications that are running on linux distributions in data centers all around the world. It seems to me that devOps folks and sysadmins benefit from having a general understanding of what the applications that are running on their servers are about. That's why I posted the research paper about training deep learning neural natworks on a small COTS HPC network. Deep learning is inextricably linked with linux in the data center. And deep learning is nothing more than advanced math and algorithms. Just my opinion. Thanks for your comments. It's good to hear from you. enjoy, Karen -- Karen Shaeffer Be aware: If you see an obstacle in your path, Neuralscape Services that obstacle is your path. Zen proverb