TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Ask HN: Why did those ideas to pool cryto mines for deep learning fail?

2 点作者 kennethjiang将近 6 年前
With cryto currency price hovering below profitable lines for most crypto mines, it seems to make so much business sense to pool these GPU resources for deep learning.<p>There have been quite a few attempts at it, such as https:&#x2F;&#x2F;aicrypto.ai&#x2F;, https:&#x2F;&#x2F;vast.ai&#x2F;, and https:&#x2F;&#x2F;www.reddit.com&#x2F;r&#x2F;gpumining&#x2F;comments&#x2F;86ofw2&#x2F;rent_out_your_gpu_compute_to_ai_researchers_and&#x2F;. However none of them seems to have gone anywhere.<p>Why?

3 条评论

duxup将近 6 年前
Is it profitable for a miner to rent out time on their mining systems?<p>The reddit thread says &quot;double what he was making by mining Ethereum, and in return he would let me train my neural networks on his computer at significantly less than what I would have had to pay AWS.&quot; That is pretty significant but I&#x27;m kinda surprised it that much less than AWS, does AWS have a really high markup?
评论 #20790151 未加载
malux85将近 6 年前
Probably because our datasets are just as valuable as the neural network topology and weights, and as the founder of several companies that use deep learning theres no way I&#x27;d let our datasets go out to machines we don&#x27;t control
billconan将近 6 年前
Who is doing deep learning at scale? Google, fb and nvidia? Do they really care about miners’ spare GPUs. What about the risk of leaking training data to public?<p>Is model parallelization solved yet? Especially Without high speed connectivity?